Skip Navigation

Posts
16
Comments
65
Joined
3 yr. ago

Data Science

  • Introduction

    I’ve written quite a lot of free software in my life. Most of it was from scratch: projects I started myself. So I get to choose where to host them – or rather, I have to choose where to host them.

    These days, all my projects are held in Git. And mostly, I put them in ‘bare’ git repositories on my personal website.

    I don’t use any git ‘forge’ system layered on top of Git, like Gitlab or Github, which automatically makes a bug tracking database for each project, and provides a convenient button for a user to open a merge request / pull request. I just use plain Git. People can ‘git clone’ my code, and there’s a web-based browsing interface (the basic gitweb) for looking around without having to clone it at all. But that’s all the automated facilities you get.

    Occasionally this confuses people, so I thought I should write something about it.

    Discussion with the author @ https://hachyderm.io/@simontatham/114111520633445984

  • That's not an affiliate link that's an anonymous tracking link.

  • These suggestions are essentially the same as other privacy and libre focused recommendations.

  • You're right. I got lazy.

  • DFRA

  • From the article:

    DeepSeek-R1 release leaves open several questions about:

    • Data collection: How were the reasoning-specific datasets curated?
    • Model training: No training code was released by DeepSeek, so it is unknown which hyperparameters work best and how they differ across different model families and scales.
    • Scaling laws: What are the compute and data trade-offs in training reasoning models?

    These questions prompted us to launch the Open-R1 project, an initiative to systematically reconstruct DeepSeek-R1’s data and training pipeline, validate its claims, and push the boundaries of open reasoning models. By building Open-R1, we aim to provide transparency on how reinforcement learning can enhance reasoning, share reproducible insights with the open-source community, and create a foundation for future models to leverage these techniques.

    In this blog post we take a look at key ingredients behind DeepSeek-R1, which parts we plan to replicate, and how to contribute to the Open-R1 project

  • I didn't read your post correctly. Yeah, that's harassment at the very least. No better than someone screaming at a retail worker because of some corporate policies.

  • It's all about your organization's size and if the organization makes use of the Anaconda controlled defaults channel. I'm not a lawyer, but your company may be liable for some licensing fee if your company is using Anaconda's repository of binaries. You'd need to consult with an actual lawyer for more reliable assessment of your potential liability.

    Switch to using miniforge and the conda-forge channel when installing and using Conda.

  • Conda itself is outside of Anaconda, Inc's control.

  • It's better that you don't use resume driven decisions. Just do whatever you are interested in.

  • Embedded software development has dramatically advanced over the past decade. What does that mean for bare-metal programming?

    At a Glance

    • Bare-metal programming is an essential skill as it enables you to understand what your system is doing at the lowest levels.
    • Even if you spend your days working with abstraction layers, bare-metal programming will guide you should abstractions fail.
    • And bare-metal skills can provide a solid foundation for troubleshooting and debugging.

  • These two are my favorite balance of fundamentals and getting to purposeful application as quickly as possible (the first link is definitely not enough, but combined with the second she should be comfortable with the syntax and able to get basic things working):https://www.kaggle.com/learn/intro-to-programminghttps://www.kaggle.com/learn/python

    This one takes its time with fundamentals and includes some projects to put them in context of building something. It's presented on Google Colab and Jupyter notebooks: https://allendowney.github.io/ThinkPython/

    Working with GIS data means cleaning data. This one covers that and a lot of common analysis tools and techniques. But it assumes a bit of programming knowledge (Good to follow up after one of the options above): https://wesmckinney.com/book/

  • Good news from September:

    Introducing the Ghostty "Quick Terminal" feature: a terminal that drops-down based on a global keybind (also sometimes known as a Doom or Quake-style terminal). This was one of Ghostty's most requested features.

    https://x.com/mitchellh/status/1840449807095394391

  • I think that Hashimoto is using this project to iron out details that are left unaddressed due to convenience for other projects and the very low impact of any single issue Hashimoto has addressed. But much like with Apple projects, Hashimoto intends for the the end product to have greater value than the sum of the parts. Unlike Apple, it will be perfomant cross platform.

    I think the only way to evaluate a project like this is to ignore the feature comparison charts and use it to see if it really is better when those details are addressed. I have a feeling that many people will agree and most will shrug their shoulders and not give it a second look if they even gave it a first one.

    I'll be trying Ghostty out soon. I hope it's great. But I'm not expecting to be blown away.

  • He seems to target GTK based on his statement:

    "On macOS, the main GUI experience is written in Swift using AppKit and SwiftUI. The tabs are native tabs, the splits are native UI components, multi-window works as you'd expect, etc. On Linux, the GUI experience is GTK using real GTK windows and other widgets.

    Features such as error messages are not implemented with a specialized terminal view, we actually use real native UI components. The point is, while the terminal surface and core logic is cross-platform, the user interaction is all purpose-built for each operating system for a true native experience."

    https://mitchellh.com/writing/ghostty-and-useful-zig-patterns

  • Lemmy still doesn't let someone post an embedded link and picture. People don't realize that you have to include the linkin the body of the post which is annoying and intuitive, specially because when creating a new post Lemmy will allow you to fill out both form fields for link and picture but only use one.

  • I'm not sure I understand the trade offs you're choosing by deploying this way. The benefit of simplicity an speed of deployment seems clear from your write-up. But are those the most important considerations? Why or why not?

  • C Programming Language @programming.dev

    Writing a C Compiler | Build a Real Programming Language from Scratch | Nora Sandler | July 2024 | No Starch Press | 792 pages | ISBN-13: 9781718500426

    nostarch.com /writing-c-compiler
  • Rust @programming.dev

    A table of publicly available Arena crates and their features

    donsz.nl /blog/arenas/
  • Rust @programming.dev

    Understanding Rust's Trait Objects: Vtables, Dynamic Dispatch, and Memory Deallocation

    www.eventhelix.com /rust/rust-to-assembly-tail-call-via-vtable-and-box-trait-free/
  • Raspberry Pi @programming.dev

    Why you should fall in love with the RP2350 | Dmitry Grinberg | Aug 8, 2024

    dmitry.gr
  • Rust @programming.dev

    System76 with Jeremy Soller | Rust in Production Podcast S02 E07 by corrode Rust Consulting | 2024-07-25

    corrode.dev /podcast/s02e07-system76/
  • Rust @programming.dev

    Rust has a HUGE supply chain security problem | Sylvain Kerkour | July 2, 2024

    kerkour.com /rust-supply-chain-security-standard-library
  • Rust @programming.dev

    This Month in Redox - March 2024 - Redox - Your Next(Gen) OS

    www.redox-os.org /news/this-month-240330
  • Rust @programming.dev

    Pre-Pooping Your Pants With Rust

    cglab.ca /~abeinges/blah/everyone-poops/
  • Rust @programming.dev

    Rust Atomics and Locks by Mara Bos

    marabos.nl /atomics/
  • Rust @programming.dev

    What are you all doing with Rust?

  • Programming @programming.dev

    Do Users Write More Insecure Code with AI Assistants?

    chaos.social /@nblr/111698366167829445
  • Programming @programming.dev

    Japan determines copyright doesn't apply to LLM/ML training data

    infosec.town /notes/9o2c3aijben6rgxe
  • Linux @programming.dev

    Nala: a frontend for APT, so you can have prettier output, faster downloads of packages, and a history.

    linuxiac.com /nala-apt-command-frontend/
  • Programming @programming.dev

    What I Learned from Maintaining a Repo During Hacktoberfest and Merging 356 PRs (November 2, 2022)

    www.freecodecamp.org /news/what-i-learned-as-a-hacktoberfest-repo-maintainer/
  • Selfhosted @lemmy.world

    This should be a pinned post as it really captures the essence of my experience so far.

  • Programming @programming.dev

    mastodon.social /@objc/110894496445183461