tinygrad
root
tinygrad | root | |
---|---|---|
17 | 31 | |
24,018 | 2,421 | |
3.3% | 1.2% | |
10.0 | 10.0 | |
6 days ago | 6 days ago | |
Python | C++ | |
MIT License | GNU General Public License v3.0 or later |
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.
tinygrad
-
AMD Unveils Ryzen 8000G Series Processors: Zen 4 APUs for Desktop with Ryzen AI
Not sure if I completely understand what "Ryzen AI" does, but Tinygrad for example has some limited support for RDNA3[0]. It isn't quite there yet in matters of performance though, as you can read in the comments of that file.
There's also a small tutorial by AMD on how to use the WMMA intrinsic[1] using AMD's hipcc[2] compiler. Documentation is sparse kinda sparse, but the instruction set is not huge. The RDNA3 ISA guide[3] might also be helpful (and only a fraction of the pages are relevant.)
0. https://github.com/tinygrad/tinygrad/blob/master/extra/gemm/...
1. https://gpuopen.com/learn/wmma_on_rdna3/
2. https://github.com/ROCm/HIPCC
3. https://www.amd.com/content/dam/amd/en/documents/radeon-tech...
- Tinygrad 0.8.0 Release
-
Beyond Backpropagation - Higher Order, Forward and Reverse-mode Automatic Differentiation for Tensorken
This post describes how I added automatic differentiation to Tensorken. Tensorken is my attempt to build a fully featured yet easy-to-understand and hackable implementation of a deep learning library in Rust. It takes inspiration from the likes of PyTorch, Tinygrad, and JAX.
-
[D] What is a good way to maintain code readability and code quality while scaling up complexity in libraries like Hugging Face?
what do you think about tinygrad? I think its a good example of growing and well written, (partially) well documented library with many close to reference implementations
-
AMD MI300 Performance – Faster Than H100, but How Much?
The idea of model architecture making fast hardware design easier is what makes https://github.com/tinygrad/tinygrad so interesting.
-
💻 7 Open-Source DevTools That Save Time You Didn't Know to Exist ⌛🚀
🌟 Support on GitHub Website: https://tinygrad.org/
- Tinygrad
-
How to train an Iris dataset classifier with Tinygrad
Before we begin, make sure you have TinyGrad and the required dependencies installed. You can find the installation instructions here.
-
Decomposing Language Models into Understandable Components
Try to get something like tinygrad[1] running locally, that way you can tweak things a bit run it again and see how it performs. While doing this you'll pick up most of the concepts and get a feeling of how things work. Also, take a look at projects like llama.cpp[2], you don't have to fully understand what's going on here, tho.
You may need some intermediate knowledge of linear algebra and this thing called "data science" nowadays, which is pretty much knowing how to mangle data and visualize it.
Try creating a small model on your own, it doesn't have to be super fancy just make sure it does something you want it to do. And then ... you'll probably could go on your own then.
1: https://github.com/tinygrad/tinygrad
2: https://github.com/ggerganov/llama.cpp
- Tinygrad 0.7.0
root
-
If you can't reproduce the model then it's not open-source
I think the process of data acquisition isn't so clear-cut. Take CERN as an example: they release loads of data from various experiments under the CC0 license [1]. This isn't just a few small datasets for classroom use; we're talking big-league data, like the entire first run data from LHCb [2].
On their portal, they don't just dump the data and leave you to it. They've got guides on analysis and the necessary tools (mostly open source stuff like ROOT [3] and even VMs). This means anyone can dive in. You could potentially discover something new or build on existing experiment analyses. This setup, with open data and tools, ticks the boxes for reproducibility. But does it mean people need to recreate the data themselves?
Ideally, yeah, but realistically, while you could theoretically rebuild the LHC (since most technical details are public), it would take an army of skilled people, billions of dollars, and years to do it.
This contrasts with open source models, where you can retrain models using data to get the weights. But getting hold of the data and the cost to reproduce the weights is usually prohibitive. I get that CERN's approach might seem to counter this, but remember, they're not releasing raw data (which is mostly noise), but a more refined version. Try downloading several petabytes of raw data if not; good luck with that. But for training something like a LLM, you might need the whole dataset, which in many cases have its own problems with copyrights…etc.
[1] https://opendata.cern.ch/docs/terms-of-use
[2] https://opendata.cern.ch/docs/lhcb-releases-entire-run1-data...
[3] https://root.cern/
- What software is used to generate plots/graphs like this seen in many particle physics papers?
-
Interactive GCC (igcc) is a read-eval-print loop (REPL) for C/C++
The odd part is that this is not just for fun. For many physicists when I was at CERN, a C++ REPL was a commonly used tool to interactively debug analyses to such a degree that many never compiled their code. Back then, I believe, it was some custom implementation included in ROOT (https://root.cern/). I even went out of my way to write C++ code compatible to it just so it could run with this implementation, otherwise some colleagues weren't interested in collaborating at all.
-
Stable Diffusion in pure C/C++
That Python ML code is calling C++ code running in the GPU, one more reason to use C++ across the whole stack.
CERN already used prototyping in C++, with ROOT and CINT, 20 years ago.
https://root.cern/
Nowadays it is even usable from Netbooks via Xeus.
It is more a matter of lack of exposure to C++ interpreters than anything else.
- Root: Analyzing Petabytes of Data, Scientifically
-
Aliens might be waiting for humans to solve a puzzle
Quantum computing is a pretty interesting science too. https://home.cern/news/press-release/knowledge-sharing/cern-quantum-technology-initiative-unveils-strategic-roadmap they have to deal with lots of data streaming too https://root.cern/
-
cppyy Generated Wrappers and Type Annotations
I'm a user of CERN's ROOT (https://root.cern/) and while I'd usually write in C++, I've been trying to write as much Python as I can recently to get a bit better in the language.
- Root: Analyzing Petabytes of Scientific Data
-
Span: how to cast pointer of pointer to other types?
I'm dealing with a C++ software called ROOT made by CERN, which is, if I'm not wrong, the only C++ API that we could use for data analysis such as plotting histograms, fitting multi-parameter functions and storing data in the size of TB to the disk and many more. That's the only reason why physicists still stick to this software. you can check here .
-
How exactly would you go about writing a program to simplify algebraic expressions?
Hey, I found something which could be useful: https://root.cern
What are some alternatives?
Pytorch - Tensors and Dynamic neural networks in Python with strong GPU acceleration
PyMesh - Geometry Processing Library for Python
jax - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
xeus - Implementation of the Jupyter kernel protocol in C++
llama.cpp - LLM inference in C/C++
tfgo - Tensorflow + Go, the gopher way
llama - Inference code for Llama models
windows-telemetry-blocklist - Blocks outgoing Windows telemetry, compatible with Pi-Hole.
openpilot - openpilot is an open source driver assistance system. openpilot performs the functions of Automated Lane Centering and Adaptive Cruise Control for 250+ supported car makes and models.
decimal - Arbitrary-precision fixed-point decimal numbers in Go
tensorflow_macos - TensorFlow for macOS 11.0+ accelerated using Apple's ML Compute framework.
apd - Arbitrary-precision decimals for Go