mirror of
https://github.com/discourse/discourse-ai.git
synced 2025-10-31 06:28:48 +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>
36 lines
1.1 KiB
Ruby
36 lines
1.1 KiB
Ruby
# frozen_string_literal: true
|
|
|
|
RSpec.describe DiscourseAi::Embeddings::Strategies::Truncation do
|
|
subject(:truncation) { described_class.new }
|
|
|
|
describe "#prepare_text_from" do
|
|
context "when using vector from OpenAI" do
|
|
before { SiteSetting.max_post_length = 100_000 }
|
|
|
|
fab!(:topic) { Fabricate(:topic) }
|
|
fab!(:post) do
|
|
Fabricate(:post, topic: topic, raw: "Baby, bird, bird, bird\nBird is the word\n" * 500)
|
|
end
|
|
fab!(:post) do
|
|
Fabricate(
|
|
:post,
|
|
topic: topic,
|
|
raw: "Don't you know about the bird?\nEverybody knows that the bird is a word\n" * 400,
|
|
)
|
|
end
|
|
fab!(:post) { Fabricate(:post, topic: topic, raw: "Surfin' bird\n" * 800) }
|
|
|
|
let(:model) do
|
|
DiscourseAi::Embeddings::VectorRepresentations::TextEmbeddingAda002.new(truncation)
|
|
end
|
|
|
|
it "truncates a topic" do
|
|
prepared_text =
|
|
truncation.prepare_text_from(topic, model.tokenizer, model.max_sequence_length)
|
|
|
|
expect(model.tokenizer.size(prepared_text)).to be <= model.max_sequence_length
|
|
end
|
|
end
|
|
end
|
|
end
|