neurallambda
neurallambda
In the PID example, there's a line (https://github.com/mstksg/auto-examples/blob/master/src/Recursive.hs#L103): ```haskell control
Thanks for the help on Stackoverflow today! I've really been enjoying getting to know your library, and in conjunction, learning tradeoffs between your approach and that of Conal Elliot in...
Transformers can now return `past_key_values` which can be used to speed up future calls: https://github.com/huggingface/transformers/pull/25086 I've noticed in experiments that when I build a query function in LMQL, I _thought_...
With this query, the output can include many newlines (Using Falcon 7B): ```python @lmql.query() def user(): '''lmql '[USER]' where STOPS_AT(USER, "\n") and len(TOKENS(USER)) > 100 ''' ``` I would expect...
Model args are not processed correctly when invoked like eg: ``` lmql.model("/home/me/models/phi-1_5", cuda=True, trust_remote_code=True, dtype='float16') ``` **Reproduction:** Fire up an `lmql` server: ``` $ lmql serve-model "distilgpt2" --cuda ``` Try...
I need to be able to control the lifecycle of models in VRAM to work on smaller devices. My first approach was to use your `local:the_model` syntax. At some pain,...
Hi, I'm the author of [UniteAI](https://github.com/freckletonj/uniteai), and I like what you're working on. My project is similar but I like how you've designed the client interface. The major difference is,...
Huggingface transformers offers `stopping_criteria`: https://huggingface.co/transformers/v4.6.0/_modules/transformers/generation_stopping_criteria.html I use this with a `threading.Event` so I can stop generation from a separate thread, and it works great with `transformers`: ```python def custom_stopping_criteria(local_llm_stop_event): def...
> I just tried the uniteai-lsp in emacs. I got it to work, but it did require some tinkering. Installing uniteai[all] couldn't build 'pyaudio', and installing uniteai[openai] did not have...