memorization
hydra-zen
memorization | hydra-zen | |
---|---|---|
1 | 1 | |
5 | 284 | |
- | 5.3% | |
10.0 | 9.4 | |
over 2 years ago | 4 days ago | |
Python | Python | |
MIT License | MIT License |
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.
memorization
-
[D] DALL·E to be made available as API, OpenAI to give users full ownership rights to generated images
Codex is not technically copy pasting; it is generating a new output that is (almost) exactly the same, or indistinguishable on the eyes of a human, to the input. Sounds like semantics, but there is no actual copying. You already have music generating algorithms that can also generate short samples that are indistinguishable to the inputs (memorisation). Dall-E 2 is not there yet, but we are close to prompting "Original Mona Lisa painting" and be given back the original Mona Lisa painting with striking similarities. There are already several generative models of images that can mostly memorise inputs used to train it (quick example found using google: https://github.com/alan-turing-institute/memorization).
hydra-zen
-
[Project] I built a minimal stateless ML project template built on my current favourite stack
It provides mature configuration support via [Hydra-Zen](https://github.com/mit-ll-responsible-ai/hydra-zen) and automates configuration generation via [decorators](https://github.com/BayesWatch/minimal-ml-template/blob/af387e59472ea67552b4bb8972b39fe95952dd8a/mlproject/decorators.py#L10) implemented in this repo.
What are some alternatives?
torch-fidelity - High-fidelity performance metrics for generative models in PyTorch
kubeproject - A set of tools built to simplify daily driving of cloud resources for individual VM access, Kubernetes batch jobs and miscellaneous useful functionality related to cloud-based ML research
wandb - 🔥 A tool for visualizing and tracking your machine learning experiments. This repo contains the CLI and Python API.
traingenerator - 🧙 A web app to generate template code for machine learning
Sacred - Sacred is a tool to help you configure, organize, log and reproduce experiments developed at IDSIA.
forward-forward-pytorch - Forward Forward algorithm (by Geoffrey Hinton) implemented with pytorch.
benchmark_VAE - Unifying Variational Autoencoder (VAE) implementations in Pytorch (NeurIPS 2022)
minimal-ml-template - A very minimal ml project template that uses HF transformers and wandb to train a simple NN and evaluate it, in a stateless manner compatible with Spot instances kubernetes workflows
disentangling-vae - Experiments for understanding disentanglement in VAE latent representations
lightning-hydra-template - PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡