Rename `model_name` and `model_name_or_path` parameters as `model`

deepset-ai / haystack

:mag: AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.

https://haystack.deepset.ai

Apache License 2.0

17.65k stars 1.91k forks source link

Rename `model_name` and `model_name_or_path` parameters as `model` #6534

Closed bilgeyucel closed 10 months ago

bilgeyucel commented 11 months ago

Is your feature request related to a problem? Please describe. As a user, I'd like to have a similar interface for each Generator (possibly for all components accepting a model name) to pass the model name

Describe the solution you'd like We have discussed that parameters should be named as model but the most important point is to have consistency

Describe alternatives you've considered Having more documentation around the parameter names can be another solution

Additional context N/A

bilgeyucel commented 10 months ago

If there's nothing against it, it would be nice to have the same convention for all components including embedders and transcribers

ZanSara commented 10 months ago

Related PRs:

[x] https://github.com/deepset-ai/haystack/pull/6715
[x] https://github.com/deepset-ai/haystack/pull/6733
[x] https://github.com/deepset-ai/haystack/pull/6731
[x] https://github.com/deepset-ai/haystack/pull/6734
[x] https://github.com/deepset-ai/haystack/pull/6736
[x] https://github.com/deepset-ai/haystack/pull/6744
[x] Review the documentation, docstrings, etc

ZanSara commented 10 months ago

For the related work on the integrations, see https://github.com/deepset-ai/haystack-core-integrations/issues/227