Class: Ask::RAG::TextSplitter::Base
- Inherits:
-
Object
- Object
- Ask::RAG::TextSplitter::Base
- Defined in:
- lib/ask/rag/text_splitter/base.rb
Direct Known Subclasses
Instance Method Summary collapse
-
#initialize(chunk_size: 1000, chunk_overlap: 200, length_function: :length.to_proc) ⇒ Base
constructor
A new instance of Base.
-
#split_documents(documents) ⇒ Array<Ask::Document>
Split documents into smaller documents.
-
#split_text(text) ⇒ Array<String>
Split a single string into chunks.
Constructor Details
#initialize(chunk_size: 1000, chunk_overlap: 200, length_function: :length.to_proc) ⇒ Base
Returns a new instance of Base.
10 11 12 13 14 |
# File 'lib/ask/rag/text_splitter/base.rb', line 10 def initialize(chunk_size: 1000, chunk_overlap: 200, length_function: :length.to_proc) @chunk_size = chunk_size @chunk_overlap = chunk_overlap @length_function = length_function end |
Instance Method Details
#split_documents(documents) ⇒ Array<Ask::Document>
Split documents into smaller documents.
26 27 28 29 30 31 32 33 34 35 36 |
# File 'lib/ask/rag/text_splitter/base.rb', line 26 def split_documents(documents) documents.flat_map do |doc| chunks = split_text(doc.content) chunks.map.with_index do |chunk, idx| Ask::Document.new( content: chunk, metadata: doc..merge(chunk: idx) ) end end end |
#split_text(text) ⇒ Array<String>
Split a single string into chunks.
19 20 21 |
# File 'lib/ask/rag/text_splitter/base.rb', line 19 def split_text(text) raise NotImplementedError end |