Rémi Louf
Rémi Louf
```python answer = generator( model.apply_chat_template(prompt) ) ``` It would _substantially_ simplify the code in this PR as well.
Indeed, we are not going to make it the default behavior. Users should be able to inspect the modified prompt before they pass it to the generator.
@leloykun are you still working on this?
We should do this once #1010 has been merged and then release `0.1.0`
The idea is interesting. Do we really want to increase the surface area of the library? There's a risk of spreading ourselves too thinly by adding extra code to maintain.
Yes you can simply ignore them by adding them to [this list in `pyproject.toml`](https://github.com/dottxt-ai/outlines/blob/main/pyproject.toml#L109)
These are the two entry points indeed.
That's correct, but we try to add other providers so people can compare their output to guided generation in outlines. I often need to do this.
The team has very limited resources
I can get on board with this. To follow `ast.NodeVisitor`'s naming scheme we could name the handlers `visit_X`. I think we should first implement a first version of the converter...