mirror of
https://github.com/discourse/discourse-ai.git
synced 2025-11-02 07:28:38 +00:00
2c0f535bab
* FEATURE: HyDE-powered semantic search. It relies on the new outlet added on discourse/discourse#23390 to display semantic search results in an unobtrusive way. We'll use a HyDE-backed approach for semantic search, which consists on generating an hypothetical document from a given keywords, which gets transformed into a vector and used in a asymmetric similarity topic search. This PR also reorganizes the internals to have less moving parts, maintaining one hierarchy of DAOish classes for vector-related operations like transformations and querying. Completions and vectors created by HyDE will remain cached on Redis for now, but we could later use Postgres instead. * Missing translation and rate limiting --------- Co-authored-by: Roman Rizzi <rizziromanalejandro@gmail.com>
31 lines
832 B
Ruby
31 lines
832 B
Ruby
# frozen_string_literal: true
|
|
|
|
module DiscourseAi
|
|
module Embeddings
|
|
module HydeGenerators
|
|
class OpenAi < DiscourseAi::Embeddings::HydeGenerators::Base
|
|
def prompt(search_term)
|
|
[
|
|
{
|
|
role: "system",
|
|
content: "You are a helpful bot. You create forum posts about a given topic.",
|
|
},
|
|
{ role: "user", content: "Create a forum post about the topic: #{search_term}" },
|
|
]
|
|
end
|
|
|
|
def models
|
|
%w[gpt-3.5-turbo gpt-4]
|
|
end
|
|
|
|
def hypothetical_post_from(query)
|
|
::DiscourseAi::Inference::OpenAiCompletions.perform!(
|
|
prompt(query),
|
|
SiteSetting.ai_embeddings_semantic_search_hyde_model,
|
|
).dig(:choices, 0, :message, :content)
|
|
end
|
|
end
|
|
end
|
|
end
|
|
end
|