Machine learning and other gibberish
See also: https://sharing.leima.is
Archives: https://datumorphism.leima.is/amneumarkt/
See also: https://sharing.leima.is
Archives: https://datumorphism.leima.is/amneumarkt/
#ai
Looks like we don't need "Pydantic is all you need" anymore. It's easier to built actionable models/workflows with this new update.
https://openai.com/index/introducing-structured-outputs-in-the-api/
Looks like we don't need "Pydantic is all you need" anymore. It's easier to built actionable models/workflows with this new update.
https://openai.com/index/introducing-structured-outputs-in-the-api/
Openai
Introducing Structured Outputs in the API
We are introducing Structured Outputs in the API—model outputs now reliably adhere to developer-supplied JSON Schemas.
#tools
Looks like github is providing foundation models soon.
Marketplace
https://github.com/marketplace/models
Looks like github is providing foundation models soon.
Marketplace
https://github.com/marketplace/models
GitHub
GitHub is where people build software. More than 150 million people use GitHub to discover, fork, and contribute to over 420 million projects.
#math
‘Sensational breakthrough’ marks step toward revealing hidden structure of prime numbers | Science | AAAS
https://www.science.org/content/article/sensational-breakthrough-marks-step-toward-revealing-hidden-structure-prime-numbers
‘Sensational breakthrough’ marks step toward revealing hidden structure of prime numbers | Science | AAAS
https://www.science.org/content/article/sensational-breakthrough-marks-step-toward-revealing-hidden-structure-prime-numbers
Science
‘Sensational breakthrough’ marks step toward revealing hidden structure of prime numbers
Proof provides fresh tools to attack famed Riemann hypothesis, math’s biggest unsolved problem
#ml
Meta's second version of segment anything.
https://github.com/facebookresearch/segment-anything-2
They have a nice demo:
https://sam2.metademolab.com/
Meta's second version of segment anything.
https://github.com/facebookresearch/segment-anything-2
They have a nice demo:
https://sam2.metademolab.com/
GitHub
GitHub - facebookresearch/sam2: The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM…
The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th...
#ai
Deep learning model quantization:
A Visual Guide to Quantization - by Maarten Grootendorst
https://newsletter.maartengrootendorst.com/p/a-visual-guide-to-quantization#%C2%A7symmetric-quantization
Deep learning model quantization:
A Visual Guide to Quantization - by Maarten Grootendorst
https://newsletter.maartengrootendorst.com/p/a-visual-guide-to-quantization#%C2%A7symmetric-quantization
Maartengrootendorst
A Visual Guide to Quantization
Exploring memory-efficient techniques for LLMs
#ai
123B
https://mistral.ai/news/mistral-large-2407/
123B
https://mistral.ai/news/mistral-large-2407/
mistral.ai
Large Enough | Mistral AI
Today, we are announcing Mistral Large 2, the new generation of our flagship model. Compared to its predecessor, Mistral Large 2 is significantly more capable in code generation, mathematics, and reasoning. It also provides a much stronger multilingual support…
RAG is evolving so fast.
Gao, Yunfan, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, Meng Wang, and Haofen Wang. 2023. “Retrieval-Augmented Generation for Large Language Models: A Survey.” arXiv [Cs.CL]. arXiv. http://arxiv.org/abs/2312.10997.
#dl
There is this new lib called scale. One could compile CUDA code to use it on AMD GPU.
https://docs.scale-lang.com/manual/how-to-use/
I don't know who is more pissed off, NVidia or AMD.
There is this new lib called scale. One could compile CUDA code to use it on AMD GPU.
https://docs.scale-lang.com/manual/how-to-use/
I don't know who is more pissed off, NVidia or AMD.
#misc
A podcast episode about the random bill thief.
https://open.spotify.com/episode/5M4ghNl16ZA3Cnx41KWQRy?si=DJKtuwXfTL6c37BuyCfNbQ
This is the story:
https://www.nytimes.com/2019/03/25/business/facebook-google-wire-fraud.html
A podcast episode about the random bill thief.
https://open.spotify.com/episode/5M4ghNl16ZA3Cnx41KWQRy?si=DJKtuwXfTL6c37BuyCfNbQ
This is the story:
https://www.nytimes.com/2019/03/25/business/facebook-google-wire-fraud.html
NY Times
He Tried to Bilk Google and Facebook Out of $100 Million With Fake Invoices (Published 2019)
A Lithuanian man pleaded guilty last week to wire fraud and said he was part of a scheme to imitate vendors and fraudulently invoice corporations.
#ai
https://github.com/postgresml/korvus
https://github.com/postgresml/korvus
GitHub
GitHub - postgresml/korvus: Korvus is a search SDK that unifies the entire RAG pipeline in a single database query. Built on top…
Korvus is a search SDK that unifies the entire RAG pipeline in a single database query. Built on top of Postgres with bindings for Python, JavaScript, Rust and C. - postgresml/korvus
#fun
Not sure how they draw that conclusion though.
https://www.reddit.com/r/dataengineering/s/xw5XAu5JQ4
Not sure how they draw that conclusion though.
https://www.reddit.com/r/dataengineering/s/xw5XAu5JQ4
Reddit
From the dataengineering community on Reddit
Explore this post and more from the dataengineering community
#ml
I was searching for a tool to visualize computational graphs and ran into this preprint. The hierarchical visualization idea is quite nice.
https://arxiv.org/abs/2212.10774
I was searching for a tool to visualize computational graphs and ran into this preprint. The hierarchical visualization idea is quite nice.
https://arxiv.org/abs/2212.10774
#ml
Schmidhuber J. Deep Learning: Our Miraculous Year 1990-1991. In: arXiv.org [Internet]. 12 May 2020 [cited 7 Jul 2024]. Available: https://arxiv.org/abs/2005.05744
Schmidhuber J. Deep Learning: Our Miraculous Year 1990-1991. In: arXiv.org [Internet]. 12 May 2020 [cited 7 Jul 2024]. Available: https://arxiv.org/abs/2005.05744
#visualization
Beautiful. There are details that the author designed quite well.
How long are the nights?
https://plotparade.com/gallery_sunrise.html
Beautiful. There are details that the author designed quite well.
How long are the nights?
https://plotparade.com/gallery_sunrise.html
#statistics
repository compiled of seminal papers in the field of statistics
https://ledaliang.github.io/journalclub/
repository compiled of seminal papers in the field of statistics
https://ledaliang.github.io/journalclub/
