RWKV-LM

deep-learning

RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

How to download and setup RWKV-LM

Open terminal and run command

git clone https://github.com/BlinkDL/RWKV-LM.git

git clone is used to create a copy or clone of RWKV-LM repositories. You pass git clone a repository URL.
it supports a few different network protocols and corresponding URL formats.

Also you may download zip file with RWKV-LM https://github.com/BlinkDL/RWKV-LM/archive/master.zip

Or simply clone RWKV-LM with SSH

[email protected]:BlinkDL/RWKV-LM.git

If you have some problems with RWKV-LM

You may open issue on RWKV-LM support forum (system) here: https://github.com/BlinkDL/RWKV-LM/issues

Similar to RWKV-LM repositories

Here you may see RWKV-LM alternatives and analogs

tensorflow keras TensorFlow-Examples pytorch CNTK data-science-ipython-notebooks Qix handong1587.github.io telegram-list machine-learning-curriculum caffe machine-learning-for-software-engineers awesome-deep-learning-papers incubator-mxnet lectures cs-video-courses Screenshot-to-code spaCy cheatsheets-ai awesome-deep-learning handson-ml ML-From-Scratch dive-into-machine-learning tfjs tflearn awesome-datascience openpose tfjs-core Machine-Learning-Tutorials EffectiveTensorflow