apoorva lal
@apoorvalal.com
causal inference, econometrics, ML, arsenal, loud music, unix, FOSS for scientific computing.
apoorvalal.github.io
(passively) maintains @paperposterbot.bsky.social
apoorvalal.github.io
(passively) maintains @paperposterbot.bsky.social
google scholar pdf reader's citation popups are sending me down interesting rabbit-holes. good job goog.
November 4, 2025 at 7:43 PM
google scholar pdf reader's citation popups are sending me down interesting rabbit-holes. good job goog.
yeah i suspect they'll be quite parsimonious actually, you could even try a few simple options in spacy and spot-check performance before going for a big model off huggingface. unclear if an LLM running on top will add value but reasonable people seem to disagree with me and put everything in chat
November 3, 2025 at 10:08 PM
yeah i suspect they'll be quite parsimonious actually, you could even try a few simple options in spacy and spot-check performance before going for a big model off huggingface. unclear if an LLM running on top will add value but reasonable people seem to disagree with me and put everything in chat
was going to apologise for giving the gift horse a dental examination but since your other pasttime is a genuinely masochistic game i think debugging cuda issues is preferable bsky.app/profile/paul...
F Bilewater
important PSA for everyone still struggling their way through silksong: you are missing a bench in bilewater
www.ign.com/articles/hol...
www.ign.com/articles/hol...
November 3, 2025 at 9:11 PM
was going to apologise for giving the gift horse a dental examination but since your other pasttime is a genuinely masochistic game i think debugging cuda issues is preferable bsky.app/profile/paul...
assuming you went ahead with postgres, i think you could generate embeddings with some reasonable model (openai, or huggingface.co/nomic-ai/nom...), insert it into your db, and then use pgvector to do nearest neighbours instead of a string search?
github.com/pgvector/pgv...
would be cool to have
github.com/pgvector/pgv...
would be cool to have
GitHub - pgvector/pgvector: Open-source vector similarity search for Postgres
Open-source vector similarity search for Postgres. Contribute to pgvector/pgvector development by creating an account on GitHub.
github.com
November 3, 2025 at 9:07 PM
assuming you went ahead with postgres, i think you could generate embeddings with some reasonable model (openai, or huggingface.co/nomic-ai/nom...), insert it into your db, and then use pgvector to do nearest neighbours instead of a string search?
github.com/pgvector/pgv...
would be cool to have
github.com/pgvector/pgv...
would be cool to have
arxiv.org/abs/2307.11255 i learned a lot from this review fwiw
A Framework for Statistical Inference via Randomized Algorithms
Randomized algorithms, such as randomized sketching or stochastic optimization, are a promising approach to ease the computational burden in analyzing large datasets. However, randomized algorithms al...
arxiv.org
November 2, 2025 at 5:24 PM
arxiv.org/abs/2307.11255 i learned a lot from this review fwiw
applied econometrics basically restricts itself to this by sticking to stacking heaps of least squares and (with a few exceptions) has basically missed the boat on the last 40 years of computational advancements.
November 2, 2025 at 5:21 PM
applied econometrics basically restricts itself to this by sticking to stacking heaps of least squares and (with a few exceptions) has basically missed the boat on the last 40 years of computational advancements.
Lol to stretch the metaphor beyond breaking point: twine is used to move the wheels? Cut a chunk off a wheel of cheese?
twine.readthedocs.io/en/stable/
twine.readthedocs.io/en/stable/
November 1, 2025 at 4:38 PM
Lol to stretch the metaphor beyond breaking point: twine is used to move the wheels? Cut a chunk off a wheel of cheese?
twine.readthedocs.io/en/stable/
twine.readthedocs.io/en/stable/
github.com/py-econometr... i've had to explain why the regression library that goes fast because of ducks has this logo
GitHub - py-econometrics/duckreg: Every big regression is a small regression with weights.
Every big regression is a small regression with weights. - GitHub - py-econometrics/duckreg: Every big regression is a small regression with weights.
github.com
October 31, 2025 at 5:55 PM
github.com/py-econometr... i've had to explain why the regression library that goes fast because of ducks has this logo
I use "scikit-learning" (complimentary)
October 31, 2025 at 1:49 PM
I use "scikit-learning" (complimentary)
not sure why What we do in the shadows hasn't crobarred in a set-piece here
October 30, 2025 at 6:10 PM
not sure why What we do in the shadows hasn't crobarred in a set-piece here
Careful, ridley scott has run out of ideas and might retcon this into the prometheus prequel timeline
October 27, 2025 at 5:13 AM
Careful, ridley scott has run out of ideas and might retcon this into the prometheus prequel timeline
predicting the next token.
October 26, 2025 at 6:34 PM
predicting the next token.
i don't have reason to get this but i'd also be curious if it runs on www.raspberrypi.com/products/ras... - an entire computer in a keyboard case
Buy a Raspberry Pi 500+ – Raspberry Pi
The premium desktop computer.
www.raspberrypi.com
October 26, 2025 at 3:58 PM
i don't have reason to get this but i'd also be curious if it runs on www.raspberrypi.com/products/ras... - an entire computer in a keyboard case
get an old thinkpad off ebay; T4xx series is good value and can be upgraded easily (i bumped up RAM and SSD).
October 26, 2025 at 3:32 PM
get an old thinkpad off ebay; T4xx series is good value and can be upgraded easily (i bumped up RAM and SSD).
the sequel: thank heavens for frankly ridiculous hardware (perixx periboard 624)
October 25, 2025 at 4:36 PM
the sequel: thank heavens for frankly ridiculous hardware (perixx periboard 624)
yea postgres or duckdb seems more appropriate for the scale you're probably dealing with. For small stuff, even sqlite is fine, e.g. github.com/apoorvalal/l... this reads and writes to a shopping list database that is automatically synced across devices by virtue of just being a webpage
October 25, 2025 at 1:51 AM
yea postgres or duckdb seems more appropriate for the scale you're probably dealing with. For small stuff, even sqlite is fine, e.g. github.com/apoorvalal/l... this reads and writes to a shopping list database that is automatically synced across devices by virtue of just being a webpage
wrap this in a steamlit/fasthtml webapp ; should be straightforward to vibecode a web interface to a duckdb call
bsky.app/profile/apoo... there's a repo with all the source for my applications in this thread
bsky.app/profile/apoo... there's a repo with all the source for my applications in this thread
Built and hosted a scrappy fasthtml radio player on a VPS for my own edification+excuse to listen to good radio (Soma, kexp, etc).
Curious how many concurrent users can use it before it crashes the tiny raspberry-pi tier machine i deployed this on.
lalten.org/radio
Curious how many concurrent users can use it before it crashes the tiny raspberry-pi tier machine i deployed this on.
lalten.org/radio
Lalten 🏮 Web Radio Player
lalten.org
October 25, 2025 at 1:43 AM
wrap this in a steamlit/fasthtml webapp ; should be straightforward to vibecode a web interface to a duckdb call
bsky.app/profile/apoo... there's a repo with all the source for my applications in this thread
bsky.app/profile/apoo... there's a repo with all the source for my applications in this thread
i think I saw this inscription on Ashoka's pillar in Lumbini
October 25, 2025 at 12:16 AM
i think I saw this inscription on Ashoka's pillar in Lumbini
yea this is Alan Wake 2 but with brain damage
October 24, 2025 at 4:07 PM
yea this is Alan Wake 2 but with brain damage
(not much to add on the binary outcomes piece; weird to be thwarted by character limits in 2025)
October 24, 2025 at 4:21 AM
(not much to add on the binary outcomes piece; weird to be thwarted by character limits in 2025)