mamba
devpi
Our great sponsors
mamba | devpi | |
---|---|---|
15 | 7 | |
9,307 | 823 | |
26.9% | 2.9% | |
8.3 | 9.2 | |
5 days ago | 7 days ago | |
Python | Python | |
Apache License 2.0 | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
mamba
-
Based: Simple linear attention language models
> how the recall can grow unbounded with no tradeoff
this? https://github.com/state-spaces/mamba/issues/175
-
Mamba: The Easy Way
If you want to learn this stuff as a computer engineer, you can read the code here [0]. I find the math quite helpful.
[0]: https://github.com/state-spaces/mamba
- FLaNK Stack 05 Feb 2024
- Introduction to State Space Models (SSM)
-
Fortran inference code for the Mamba state space language model
This model was discussed recently: https://news.ycombinator.com/item?id=38522428 It's a new kind of ML model architecture that can be used instead of a transformer in LLMs.
See also the original repo from the paper: https://github.com/state-spaces/mamba
-
Mamba outperforms transformers "everywhere we tried"
[2] - https://github.com/state-spaces/mamba
Out of curiosity, does anyone feel as though there's any benefit to linking to reddit when we can link to whatever the link is? I for one do not click the link and read discussion on reddit - if I wanted that sort of discussion, I would browse there, not HN.
- GitHub – State-Spaces/Mamba
-
Generate valid JSON with Mamba models
The library is compatible with any auto-regressive model, not transformers. To prove our point we integrated Mamba, a new state-space model architecture, to the library. Try it out!
-
[D] Thoughts on Mamba?
I ran the NanoGPT of Karparthy replacing Self-Attention with Mamba on his TinyShakespeare Dataset and within 5 minutes it started spitting out the following:
-
Mamba-Chat: A Chat LLM based on State Space Models
You might have come across the paper Mamba paper in the last days, which was the first attempt at scaling up state space models to 2.8B parameters to work on language data.
devpi
-
Private Python Packages With devpi
There are cases where you want the flexibility of installing a python package via pip without having it available to the open public. This article will focus on using devpi to provide a self-hosted pip compatible python package server. Ubuntu will be used for the OS as it's a fairly common Linux distribution and easily available on Windows Linux Subsystem.
-
Adding Virtual Environments to Git Repo
When not providing all dependencies yourself, you might suffer from people deleting the packages you depend on (IMHO a very rare scenario). If it is really that critical (hint: usually it isn't), create a local mirror of Pypi (full or only the packages you need). Devpi, Artifactory, etc. can do that or you just dump the necessary files into Cloud storage, so you have a backup.
-
PyPI in a Box
devpi acts as a caching proxy for PyPI and takes a bit less setup than this. Plus, you can use it for storing your own packages in a separate index.
https://github.com/devpi/devpi
-
Publishing to a private Python repository with Poetry
There are several open-source solutions for provisioning your own, personal PyPI server — for example, this can be done using pypiserver or devpi. However, configuring these services takes time and effort, and it costs money to deploy them. Instead, we’ll use Packagr, a cloud-hosted python package server that allows you to provision your own private Python package repository. It also supports NPM packages and even Docker registries and is very easy to set up. You can get started by creating a free trial account on Packagr — when you’ve created your account, you’ll see this:
-
Is there away to install Packages when not online?
devpi-server will probably for the bill
-
Python Best Practices for a New Project in 2021
> One should probably run their own package server like https://github.com/pypiserver/pypiserver
Never used pypiserver but I’ve had a good experience with https://github.com/devpi/devpi
- Devpi/devpi: Python PyPi staging server and packaging, testing, release tool
What are some alternatives?
miniforge - A conda-forge distribution.
pypiserver - Minimal PyPI server for uploading & downloading packages with pip/easy_install
pip - The Python package installer
Poetry - Python packaging and dependency management made easy
llm.f90 - LLM inference in Fortran
helm - The Kubernetes Package Manager
conda - A system-level, binary package and environment manager running on all major operating systems and platforms.
mamba-chat - Mamba-Chat: A chat LLM based on the state-space model architecture 🐍
spack - A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
warehouse - The Python Package Index