HNHacker News
TopNewBestAskShowJobs

kashifr

753 karma · joined March 11, 2011

https://github.com/kashif https://twitter.com/krasul
submissionscomments
kashifr··on DeepMind's WeatherNext model achieves breakthrough forecasting cyclones
Check out my pytorch reproduction of the paper here for those interested: https://github.com/NVIDIA/physicsnemo/pull/1660
kashifr··on TinyLoRA – Learning to Reason in 13 Parameters
You can try out TinyLoRA in PEFT main now: https://huggingface.co/docs/peft/main/en/package_reference/t...
kashifr··on DuckDB is probably the most important geospatial software of the last decade
Have you tried https://geobase.app/ they recently also had a post about duckdb integration: https://geobase.app/blog/duckdb-1-1-3
kashifr··on MaPO: A reference-free alignment technique for diffusion models
A novel and memory-friendly preference alignment method for diffusion models that does not depend on any reference model.
kashifr··on OpenHermesPreferences: Dataset of ~1M AI preferences from teknium/OpenHermes-2.5
The dataset can be used for training preference models or aligning language models through techniques like Direct Preference Optimization (DPO).
kashifr··on QLoRA 4-bit finetuning of LLMs
An efficient finetuning approach that reduces memory usage enough to finetune a 65B parameter model on a single 48GB GPU while preserving full 16-bit finetuning task performance!
kashifr··on V0.15.0 Beyond Image Generation
The title is confusing but refers to the Huggingface Diffusers v0.15 release which brings new pipelines for video and audio to diffusers, showing that diffusion is a great choice for all sorts of generative tasks.
kashifr··on StackLlama: A hands-on guide to train LlaMa with RLHF
All the steps involved in training a LlaMa model to answer questions on Stack Exchange data with RLHF.
kashifr··on HuggingFace Diffusers 0.2 with Stable Diffusion pipeline
Diffusers now supports CompVis, Stability AI and LAION's Stable Diffusion model weights.
kashifr··on Diffusers: Modular Diffusion model library from HuggingFace
A library of pretrained diffusion models across multiple modalities, such as vision and audio.
kashifr··on Asciinema – Record and share your terminal sessions
With all the terminal recordings I have seen, the content eventually ends up at the bottom. It would be awesome to have the cursor always stay in the middle... any idea if thats possible with Asciinema?
kashifr··on Level-Up Your Machine Learning
Form my own journey I would say that a good place to start for graphical models might be "Bayesian Reasoning and Machine Learning" by Barber. It's free (http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=...). I haven't read through it, but I've heard good things. However, it doesn't cover some basic things like SVM, RVM, Neural Networks...

For those I'd suggest "Pattern Recognition and Machine Learning" by Bishop. I've read throughout this and it's really well organized and thought out. For more mathematically advanced ML stuff I'd suggest "Foundations of Machine Learning" by Mohri. For a good reference for anything else I'd suggest "Machine Learning: A Probabilistic Perspective" by Murphy. For more depth on graphical models look at "Probabilistic Graphical Models: Principles and Techniques" by Koller.

On the NLP front there's the standard texts "Speech and Language Processing" by Jurafsky and "Foundations of Statistical Natural Language Processing" by Manning.

I also like "An Introduction to Statistical Learning" by James, Witten, Hastie and Tibshirani.

kashifr··on Why have CPUs been limited in frequency to around 3.5Ghz for so many years?
According to the Sandia's Cooler whiter-paper pdf: http://prod.sandia.gov/techlib/access-control.cgi/2010/10025... the limit is due to a "Thermal Brick Wall". Basically due to a lack of progress in heat exchanger technology.

I have visualized this "Brick-Wall" as a graph with data from Wikipedia pdf: http://dl.dropbox.com/u/3215373/Thermal-Brick-Wall.pdf

kashifr··on Heroku Postgres
Hi co-founder of SpacialDB here. Cloud hosted PostGIS is already available as a Heroku plugin, stand-alone and an easy to use API. Check out http://devcenter.spacialdb.com/ for documentation and you can sign up via the Heroku addons page or http://beta.spacialdb.com/
kashifr··on Samsung releases code of WebCL implementation for WebKit
Nokia also has a version of webcl here:

http://projects.developer.nokia.com/webcl