stas00 / ml-ways
ML/DL Math and Method notes
☆59Updated last year
Alternatives and similar repositories for ml-ways:
Users that are interested in ml-ways are comparing it to the libraries listed below
- Learn CUDA with PyTorch☆19Updated 2 months ago
- Repository containing awesome resources regarding Hugging Face tooling.☆46Updated last year
- Make triton easier☆47Updated 9 months ago
- ☆76Updated 8 months ago
- Experiment of using Tangent to autodiff triton☆78Updated last year
- Load compute kernels from the Hub☆107Updated this week
- This code repository contains the code used for my "Optimizing Memory Usage for Training LLMs and Vision Transformers in PyTorch" blog po…☆87Updated last year
- Collection of autoregressive model implementation☆83Updated last month
- Train, tune, and infer Bamba model☆87Updated 2 months ago
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆68Updated 10 months ago
- Just some miscellaneous utility functions / decorators / modules related to Pytorch and Accelerate to help speed up implementation of new…☆120Updated 8 months ago
- A place to store reusable transformer components of my own creation or found on the interwebs☆48Updated last week
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆59Updated 2 months ago
- Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*☆81Updated last year
- train with kittens!☆55Updated 5 months ago
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆45Updated 8 months ago
- Automatically take good care of your preemptible TPUs☆36Updated last year
- Mobile Viewer for W&B, built on top of Flutter.☆33Updated last year
- Genalog is an open source, cross-platform python package allowing generation of synthetic document images with custom degradations and te…☆42Updated last year
- Custom kernels in Triton language for accelerating LLMs☆18Updated 11 months ago
- Context Manager to profile the forward and backward times of PyTorch's nn.Module☆84Updated last year
- Utilities for Training Very Large Models☆58Updated 6 months ago
- ☆22Updated last year
- ☆43Updated last year
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆123Updated 11 months ago
- ☆98Updated 10 months ago
- Various transformers for FSDP research☆37Updated 2 years ago
- ☆48Updated last year
- Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.☆17Updated 2 weeks ago
- Cray-LM unified training and inference stack.☆21Updated 2 months ago