Class: RagEmbeddings::Database

Inherits:
Object
  • Object
show all
Defined in:
lib/rag_embeddings/database.rb

Instance Method Summary collapse

Constructor Details

#initialize(path = "embeddings.db") ⇒ Database

Returns a new instance of Database.



5
6
7
8
9
10
11
12
13
14
 
# File 'lib/rag_embeddings/database.rb', line 5

def initialize(path = "embeddings.db")
  @db = SQLite3::Database.new(path)
  @db.execute <<~SQL
    CREATE TABLE IF NOT EXISTS embeddings (
      id INTEGER PRIMARY KEY AUTOINCREMENT,
      content TEXT NOT NULL,
      embedding BLOB NOT NULL
    );
  SQL
end

Instance Method Details

#all ⇒ Object 



21
22
23
24
25
 
# File 'lib/rag_embeddings/database.rb', line 21

def all
  @db.execute("SELECT id, content, embedding FROM embeddings").map do |id, content, blob|
    [id, content, blob.unpack("f*")]
  end
end

#insert(text, embedding) ⇒ Object 



16
17
18
19
 
# File 'lib/rag_embeddings/database.rb', line 16

def insert(text, embedding)
  blob = embedding.pack("f*")
  @db.execute("INSERT INTO embeddings (content, embedding) VALUES (?, ?)", [text, blob])
end

#top_k_similar(query_text, k: 5) ⇒ Object

“Raw” search: returns the N texts most similar to the query



28
29
30
31
32
33
34
35
36
37
 
# File 'lib/rag_embeddings/database.rb', line 28

def top_k_similar(query_text, k: 5)
  query_embedding = RagEmbeddings.embed(query_text)
  query_obj = RagEmbeddings::Embedding.from_array(query_embedding)

  all.map do |id, content, emb|
    emb_obj = RagEmbeddings::Embedding.from_array(emb)
    similarity = emb_obj.cosine_similarity(query_obj)
    [id, content, similarity]
  end.sort_by { |_,_,sim| -sim }.first(k)
end