Saturday was Anchorage airport's first day this season with a high temperature at or below freezing. This is almost two weeks later than the average over the past 50 years. #akwx #weather #Climate #Anchorage @climatologist49.bsky.social
November 8, 2026 at 4:50 PM
Saturday was Anchorage airport's first day this season with a high temperature at or below freezing. This is almost two weeks later than the average over the past 50 years. #akwx #weather #Climate #Anchorage @climatologist49.bsky.social
Reams of overeager poorly cross-correlated tabular crap in teams is not an improvement
November 5, 2026 at 5:48 AM
Reams of overeager poorly cross-correlated tabular crap in teams is not an improvement
if you have a specially configured (I hesitate to say "trained" because that implies expertise it doesn't have) chatbot, it can be useful at summarising documents with a lot of tabular data because it's good at grouping similar things together without understanding what they are.
November 9, 2026 at 7:08 PM
if you have a specially configured (I hesitate to say "trained" because that implies expertise it doesn't have) chatbot, it can be useful at summarising documents with a lot of tabular data because it's good at grouping similar things together without understanding what they are.
Ah, that's where you're wrong. You also need the Solenoid Properties plugin to better enter and manage tabular data inside your frontmatter properties ;)
November 9, 2026 at 12:18 AM
Ah, that's where you're wrong. You also need the Solenoid Properties plugin to better enter and manage tabular data inside your frontmatter properties ;)
You can now run an OCR leaderboard on your own documents without publishing them. ocr-bench shares scores only.
First board: 338 20th-century government pages, 9 open-weight models. Small ones won: LightOnOCR-3 (0.8B) 1st, Falcon-OCR (270M) 2nd.
hf.co/datasets/davanstrien/ocr-board-historical-v0
First board: 338 20th-century government pages, 9 open-weight models. Small ones won: LightOnOCR-3 (0.8B) 1st, Falcon-OCR (270M) 2nd.
hf.co/datasets/davanstrien/ocr-board-historical-v0
davanstrien/ocr-board-historical-v0 · Datasets at Hugging Face
Dataset with 54 rows (Tabular, Text). Explore, preview and download it on Hugging Face.
hf.co
November 10, 2026 at 10:24 AM
You can now run an OCR leaderboard on your own documents without publishing them. ocr-bench shares scores only.
First board: 338 20th-century government pages, 9 open-weight models. Small ones won: LightOnOCR-3 (0.8B) 1st, Falcon-OCR (270M) 2nd.
hf.co/datasets/davanstrien/ocr-board-historical-v0
First board: 338 20th-century government pages, 9 open-weight models. Small ones won: LightOnOCR-3 (0.8B) 1st, Falcon-OCR (270M) 2nd.
hf.co/datasets/davanstrien/ocr-board-historical-v0
Was captivated by the tabular recipe notation a few weeks ago and so put together a simple site oriented around presenting recipes in that format. I'm a huge fan for how it condenses down a recipe; its not a universal replacement for cooking, but is my go-to day to day. justrecipes.food
November 6, 2026 at 9:47 PM
Was captivated by the tabular recipe notation a few weeks ago and so put together a simple site oriented around presenting recipes in that format. I'm a huge fan for how it condenses down a recipe; its not a universal replacement for cooking, but is my go-to day to day. justrecipes.food
A weekend pitch :)
"Yes"
- play: fast, slow, alone, with a friend, on paper, competitively
- pairs well with coffee and lasts longer ☕️
"No"
- no ads, no subscriptions, no pings, no in-app purchases
- not a mega corporation
Pay once, Unlock Pro for $7
hexsumgame.com/signup
--
#gamedev #indiedev
"Yes"
- play: fast, slow, alone, with a friend, on paper, competitively
- pairs well with coffee and lasts longer ☕️
"No"
- no ads, no subscriptions, no pings, no in-app purchases
- not a mega corporation
Pay once, Unlock Pro for $7
hexsumgame.com/signup
--
#gamedev #indiedev
November 8, 2026 at 5:36 PM
A weekend pitch :)
"Yes"
- play: fast, slow, alone, with a friend, on paper, competitively
- pairs well with coffee and lasts longer ☕️
"No"
- no ads, no subscriptions, no pings, no in-app purchases
- not a mega corporation
Pay once, Unlock Pro for $7
hexsumgame.com/signup
--
#gamedev #indiedev
"Yes"
- play: fast, slow, alone, with a friend, on paper, competitively
- pairs well with coffee and lasts longer ☕️
"No"
- no ads, no subscriptions, no pings, no in-app purchases
- not a mega corporation
Pay once, Unlock Pro for $7
hexsumgame.com/signup
--
#gamedev #indiedev
if anyone missed the Data Science Lab 🧪
the quarto is at the link below & other examples as well.
github.com/frankiethull...
the quarto is at the link below & other examples as well.
github.com/frankiethull...
November 3, 2026 at 9:45 PM
if anyone missed the Data Science Lab 🧪
the quarto is at the link below & other examples as well.
github.com/frankiethull...
the quarto is at the link below & other examples as well.
github.com/frankiethull...
StreamTab: Training-Free Temporal Link Prediction with Tabular Foundation Models
https://arxiv.org/abs/2611.09052
https://arxiv.org/abs/2611.09052
November 9, 2026 at 3:34 PM
StreamTab: Training-Free Temporal Link Prediction with Tabular Foundation Models
https://arxiv.org/abs/2611.09052
https://arxiv.org/abs/2611.09052
BAM: Bivariate Additive Models for Interpretable Tabular Prediction
Read more: https://arxiv.org/html/2611.05940v1
Read more: https://arxiv.org/html/2611.05940v1
November 9, 2026 at 5:42 PM
BAM: Bivariate Additive Models for Interpretable Tabular Prediction
Read more: https://arxiv.org/html/2611.05940v1
Read more: https://arxiv.org/html/2611.05940v1
Amazon Science: Mitra-v2: Scaling synthetic pretraining for tabular foundation models
ai-news.ypetit.net/…
#AI #AIResearch #AmazonScience
ai-news.ypetit.net/…
#AI #AIResearch #AmazonScience
November 10, 2026 at 4:18 PM
Amazon Science: Mitra-v2: Scaling synthetic pretraining for tabular foundation models
ai-news.ypetit.net/…
#AI #AIResearch #AmazonScience
ai-news.ypetit.net/…
#AI #AIResearch #AmazonScience
A tabular foundation model rivals established species distribution models from presence-only records https://www.biorxiv.org/content/10.64898/2026.11.06.765003v1
November 9, 2026 at 9:31 PM
A tabular foundation model rivals established species distribution models from presence-only records https://www.biorxiv.org/content/10.64898/2026.11.06.765003v1
A tabular foundation model rivals established species distribution models from presence-only records https://www.biorxiv.org/content/10.64898/2026.11.06.765003v1
November 9, 2026 at 9:31 PM
A tabular foundation model rivals established species distribution models from presence-only records https://www.biorxiv.org/content/10.64898/2026.11.06.765003v1
@mcnuttandrew.bsky.social reiterates his call for more work on datatypes! -- characteristics on tabular data are easy, but other data types?!
November 9, 2026 at 7:30 PM
@mcnuttandrew.bsky.social reiterates his call for more work on datatypes! -- characteristics on tabular data are easy, but other data types?!
Yandex (Russia's Google) released Yambda-5B dataset so the researchers can develop and eval recommendation models at scale.
It is one of the world's largest open datasets of its kind, it contains 4.79 billion anonymized interactions from 1 million users across 9.39 million tracks from Yandex Music.
It is one of the world's largest open datasets of its kind, it contains 4.79 billion anonymized interactions from 1 million users across 9.39 million tracks from Yandex Music.
yandex/yambda · Datasets at Hugging Face
Dataset with 5.32B rows (Tabular, Text). Explore, preview and download it on Hugging Face.
huggingface.co
November 5, 2026 at 10:47 AM
Yandex (Russia's Google) released Yambda-5B dataset so the researchers can develop and eval recommendation models at scale.
It is one of the world's largest open datasets of its kind, it contains 4.79 billion anonymized interactions from 1 million users across 9.39 million tracks from Yandex Music.
It is one of the world's largest open datasets of its kind, it contains 4.79 billion anonymized interactions from 1 million users across 9.39 million tracks from Yandex Music.
Mingqi Yang, Renqi Wu: StreamTab: Training-Free Temporal Link Prediction with Tabular Foundation Models https://arxiv.org/abs/2611.09052 https://arxiv.org/pdf/2611.09052 https://arxiv.org/html/2611.09052
November 9, 2026 at 6:43 AM
Mingqi Yang, Renqi Wu: StreamTab: Training-Free Temporal Link Prediction with Tabular Foundation Models https://arxiv.org/abs/2611.09052 https://arxiv.org/pdf/2611.09052 https://arxiv.org/html/2611.09052
Estão verificando os ballots q são chamados de provisional, o eleitor faz uso, mas eh necessário ser revisto pós eleição.
Tem estados q tem votos por correio q tb são só contabilizados quando a eleição termina, ao invés de começar a tabular assim q recebe. E tantas outras regras.
Tem estados q tem votos por correio q tb são só contabilizados quando a eleição termina, ao invés de começar a tabular assim q recebe. E tantas outras regras.
November 8, 2026 at 2:35 PM
Estão verificando os ballots q são chamados de provisional, o eleitor faz uso, mas eh necessário ser revisto pós eleição.
Tem estados q tem votos por correio q tb são só contabilizados quando a eleição termina, ao invés de começar a tabular assim q recebe. E tantas outras regras.
Tem estados q tem votos por correio q tb são só contabilizados quando a eleição termina, ao invés de começar a tabular assim q recebe. E tantas outras regras.
Stop guessing which hyperparameters to tune, and over what range.
My updated course shows you, step by step, how to optimize XGBoost and CatBoost with Optuna.
Enroll today 👇
My updated course shows you, step by step, how to optimize XGBoost and CatBoost with Optuna.
Enroll today 👇
Master Hyperparameter Optimization for Tabular Learning
Master hyperparameter tuning for linear models, random forests, and gradient-boosting trees with grid & random search, Bayesian optimization, and multi-fidelity methods.
www.trainindata.com
November 10, 2026 at 11:03 AM
Stop guessing which hyperparameters to tune, and over what range.
My updated course shows you, step by step, how to optimize XGBoost and CatBoost with Optuna.
Enroll today 👇
My updated course shows you, step by step, how to optimize XGBoost and CatBoost with Optuna.
Enroll today 👇
I've been thinking about all these ethically problematic #AI #synth firmwares and plugins.
Just future-musing: Vibe-coding firmware to your individual taste becomes normal (and ethics resolved), so the commercial race shifts to the best or coolest tabular rasa hardware to host whatever you coded.
Just future-musing: Vibe-coding firmware to your individual taste becomes normal (and ethics resolved), so the commercial race shifts to the best or coolest tabular rasa hardware to host whatever you coded.
November 5, 2026 at 3:32 PM
What games does each language read about? Arabic Wikipedia this week: Free Fire, then GTA V. Plus releases by country and new indie repos.
Free daily: https://huggingface.co/datasets/Lilambd/world-game-signals?utm_source=bluesky&utm_campaign=world-game-signals
Free daily: https://huggingface.co/datasets/Lilambd/world-game-signals?utm_source=bluesky&utm_campaign=world-game-signals
Lilambd/world-game-signals · Datasets at Hugging Face
Dataset with 1.9k rows (Tabular, Text). Explore, preview and download it on Hugging Face.
huggingface.co
November 9, 2026 at 3:47 PM
What games does each language read about? Arabic Wikipedia this week: Free Fire, then GTA V. Plus releases by country and new indie repos.
Free daily: https://huggingface.co/datasets/Lilambd/world-game-signals?utm_source=bluesky&utm_campaign=world-game-signals
Free daily: https://huggingface.co/datasets/Lilambd/world-game-signals?utm_source=bluesky&utm_campaign=world-game-signals
TIL about font-variant-numeric: tabular-nums, tab-size, and variable scoping in CSS.
November 3, 2026 at 11:42 PM
TIL about font-variant-numeric: tabular-nums, tab-size, and variable scoping in CSS.
A new AI review! saulpw/visidata ⭐4.1/5.0
VisiData is a mature, actively maintained terminal UI “data multitool” for exploring, cleaning, and transforming tabular datasets.
https://gitrated.com/saulpw/visidata
VisiData is a mature, actively maintained terminal UI “data multitool” for exploring, cleaning, and transforming tabular datasets.
https://gitrated.com/saulpw/visidata
November 10, 2026 at 9:37 PM
A new AI review! saulpw/visidata ⭐4.1/5.0
VisiData is a mature, actively maintained terminal UI “data multitool” for exploring, cleaning, and transforming tabular datasets.
https://gitrated.com/saulpw/visidata
VisiData is a mature, actively maintained terminal UI “data multitool” for exploring, cleaning, and transforming tabular datasets.
https://gitrated.com/saulpw/visidata
now that i think about it i wonder how many actually thought that i regularly use the ai overview and support it
the answer is fuck ai though
still think tabular kitty is funny
the answer is fuck ai though
still think tabular kitty is funny
November 9, 2026 at 9:38 AM
now that i think about it i wonder how many actually thought that i regularly use the ai overview and support it
the answer is fuck ai though
still think tabular kitty is funny
the answer is fuck ai though
still think tabular kitty is funny
Fun w/ AI coding - trivial effort to build web assembly (WASM) for old tools. Took my gradschool C code hack of yn00 - a fast tabular output of pairwise Ka,Ks - to rust and added full ML codeml pairwise. Runs in your browser, no webserver. hyphaltip.github.io/subopt-kaks/
kaks dN/dS
Estimate dN/dS (Ka/Ks) for coding sequences in the browser: Yang-Nielsen 2000, Nei-Gojobori 1986 and codeml-style pairwise maximum likelihood, all 27 NCBI genetic codes. Nothing leaves your computer.
hyphaltip.github.io
November 6, 2026 at 7:35 AM
Fun w/ AI coding - trivial effort to build web assembly (WASM) for old tools. Took my gradschool C code hack of yn00 - a fast tabular output of pairwise Ka,Ks - to rust and added full ML codeml pairwise. Runs in your browser, no webserver. hyphaltip.github.io/subopt-kaks/
📄 "The Balance between Nuance and Clarity: Decluttering Tabular Sequential Graphs to Counter Money Laundering" by @salomeesteves.bsky.social, Rita Costa, Louise Fallon, and Pedro Bizarro. #ieeevis
Learn more at https://ieeevis.org/year/2026/program/content/bff4980c-ea87-46f9-afdb-dfed1a404e80/
Learn more at https://ieeevis.org/year/2026/program/content/bff4980c-ea87-46f9-afdb-dfed1a404e80/
November 10, 2026 at 2:15 PM
📄 "The Balance between Nuance and Clarity: Decluttering Tabular Sequential Graphs to Counter Money Laundering" by @salomeesteves.bsky.social, Rita Costa, Louise Fallon, and Pedro Bizarro. #ieeevis
Learn more at https://ieeevis.org/year/2026/program/content/bff4980c-ea87-46f9-afdb-dfed1a404e80/
Learn more at https://ieeevis.org/year/2026/program/content/bff4980c-ea87-46f9-afdb-dfed1a404e80/