ColossalAI
pytorch-lightning
Our great sponsors
ColossalAI | pytorch-lightning | |
---|---|---|
42 | 8 | |
37,836 | 26,883 | |
3.7% | 2.0% | |
9.7 | 9.9 | |
5 days ago | 3 days ago | |
Python | Python | |
Apache License 2.0 | 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.
ColossalAI
- FLaNK AI-April 22, 2024
- Making large AI models cheaper, faster and more accessible
-
ColossalChat: An Open-Source Solution for Cloning ChatGPT with a RLHF Pipeline
> open-source a complete RLHF pipeline ... based on the LLaMA pre-trained model
I've gotten to where when I see "open source AI" I now know it's "well, except for $some_other_dependencies"
Anyway: https://scribe.rip/@yangyou_berkeley/colossalchat-an-open-so... and https://github.com/hpcaitech/ColossalAI#readme (Apache 2) can save you some medium.com heartache at least
-
Meet ColossalChat: An Open-Source AI Solution For Cloning ChatGPT With A Complete RLHF Pipeline
Quick Read: https://www.marktechpost.com/2023/04/01/meet-colossalchat-an-open-source-ai-solution-for-cloning-chatgpt-with-a-complete-rlhf-pipeline/ Github: https://github.com/hpcaitech/ColossalAI Examples: https://chat.colossalai.org/
-
A top AI researcher reportedly left Google for OpenAI after sharing concerns the company was training Bard on ChatGPT data
One of the current methods for training competing models is to have ChatGPT literally create prompt -> completion data sets. That's what was used for https://github.com/hpcaitech/ColossalAI. A model based off of the Llama weights released by facebook, then fine tuned on ChatGPT3.5 prompt + completions. So yes, there is a good chance that google is literally using ChatGPT in the training loop.
- Colossal-AI: open-source RLHF pipeline based on LLaMA pre-trained model
- ColossalChat
-
ColossalChat: An Open-Source Solution for Cloning ChatGPT with RLHF Pipeline
Here's the github from the article:
https://github.com/hpcaitech/ColossalAI
-
Open source solution replicates ChatGPT training process
The article talks about their RLHF implementation briefly. There’s details on their RLHF implementation here: https://github.com/hpcaitech/ColossalAI/blob/a619a190df71ea3...
-
how can I make my own chatGPT?
Here’s the project on GitHub: https://github.com/hpcaitech/ColossalAI
pytorch-lightning
- Lightning AI Studios – A persistent GPU cloud environment
-
Como empezar con inteligencia artificial?
https://see.stanford.edu/Course/CS229 https://lightning.ai/ https://www.youtube.com/watch?v=00s9ireCnCw&t=57s https://towardsdatascience.com/
-
Best practice for saving logits/activation values of model in PyTorch Lightning
I've been wondering on what is the recommended method of saving logits/activations using PyTorch Lightning. I've looked at Callbacks, Loggers and ModelHooks but none of the use-cases seem to be for this kind of activity (even if I were to create my own custom variants of each utility). The ModelCheckpoint Callback in its utility makes me feel like custom Callbacks would be the way to go but I'm not quite sure. This closed GitHub issue does address my issue to some extent.
- New to ML, which is easier to learn - Tensorflow or PyTorch?
- PyTorch Lightning – DL framework to train, deploy, and ship AI fast
-
We just release a complete open-source solution for accelerating Stable Diffusion pretraining and fine-tuning!
Our codebase for the diffusion models builds heavily on OpenAI's ADM codebase , lucidrains, Stable Diffusion, Lightning and Hugging Face. Thanks for open-sourcing!
-
An elegant and strong PyTorch Trainer
For lightweight use, pytorch-lightning is too heavy, and its source code will be very difficult for beginners to read, at least for me.
-
[D] Mixed Precision Training: Difference between BF16 and FP16
For the A100 GPU, theoretical performance is the same for FP16/BF16 and both rely on the same number of bits, meaning memory should be the same. However since it's quite newly added to PyTorch, performance seems to still be dependent on underlying operators used (pytorch lightning debugging in progress here).
What are some alternatives?
DeepSpeed - DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
lnd - Lightning Network Daemon ⚡️
Megatron-LM - Ongoing research training transformer models at scale
Eclair - A scala implementation of the Lightning Network.
determined - Determined is an open-source machine learning platform that simplifies distributed training, hyperparameter tuning, experiment tracking, and resource management. Works with PyTorch and TensorFlow.
mmdetection - OpenMMLab Detection Toolbox and Benchmark
fairscale - PyTorch extensions for high performance and large scale training.
composer - Supercharge Your Model Training
DeepFaceLive - Real-time face swap for PC streaming or video calls
umbrel - A beautiful home server OS for self-hosting with an app store. Buy a pre-built Umbrel Home with umbrelOS, or install on a Raspberry Pi 4, Pi 5, any Ubuntu/Debian system, or a VPS.
PaddlePaddle - PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice (『飞桨』核心框架,深度学习&机器学习高性能单机、分布式训练和跨平台部署)
Keras - Deep Learning for humans