bx_builder_chain 0.1.12 → 0.1.13

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: eb8dc7d40d37072877b376103879593077531cecc15e4487197892eea0101c0f
-  data.tar.gz: d49614a734f1add1ead4e75d71fdead7b20437a41ed1a01d85780766dd75dd81
+  metadata.gz: 38f4062b8cf5eb13c2ed44821bff90b6a29dbfbc3a39d5c38c052f19b8df7fb6
+  data.tar.gz: e8b167d59f1764d3c2f8aff7edac1d51a0990b9971e281b3d42ce531129762cc
 SHA512:
-  metadata.gz: a847b3db61c0e705b06208061288690da66d2036449be89519d146b304a657ac31e1d2bd88dcae0c2ee91d8471e768b3ef7a8576222e3884199b15b9082bbb9e
-  data.tar.gz: eae9012d1602ad2cedf44b5173edc08eda4c17c3081a1b88808989bc1eee7540b36a2a2d01839e6a01076cab4ec26829bc569cc5a5044aa0fcb07651b0bbf66f
+  metadata.gz: 57cc603b6f20f8d66270680d2243583ce1cac57d3b36f849412fdb4c8ec77dd8830579954256aa02f9693753e088ab63c1f6e3b64aed35719f66eb9fa68c5430
+  data.tar.gz: 607857292395ca97abbe6754ac5a896df79a4dfb2333e269a03be5dcdcdacfcbf9188c02d4eb09f387e66aebcfb6cce0792c39d5cbc5cff9625b9663681efa85

data/Gemfile

@@ -18,6 +18,7 @@ gem "nokogiri"
 gem "docx"
 gem "roo"
 gem "baran"
+gem "async"
 
 gem "pry"
 

data/README.md

@@ -71,7 +71,7 @@ create the llm and client
 ```ruby
 client = BxBuilderChain::Vectorsearch::Pgvector.new(
       llm: BxBuilderChain::Llm::OpenAi.new,
-      namespace: user_id # default is nil, nil is used for global public documents
+      namespace: user_id # default is nil, nil assumes global public documents
     )
 ```
 

data/bx_builder_chain.gemspec

@@ -32,4 +32,5 @@ Gem::Specification.new do |spec|
   spec.add_dependency "nokogiri", "~> 1.8" 
   spec.add_dependency "docx", "~> 0.8.0"
   spec.add_dependency "roo", "~> 2.8.3"
+  spec.add_dependency "async", "~> 1.31.0"
 end

data/lib/bx_builder_chain/vectorsearch/pgvector.rb

@@ -69,10 +69,13 @@ module BxBuilderChain::Vectorsearch
     # @return [Array<Integer>] The the ids of the added texts.
     def add_texts(texts:, ids: nil)
       if ids.nil? || ids.empty?
-        data = texts.map do |text|
-          {content: text, vectors: llm.embed(text: text).to_s, namespace: namespaces[0]}
+        Async do
+          data = texts.map do |text|
+            Async do
+              {content: text, vectors: llm.embed(text: text).to_s, namespace: namespaces[0]}
+            end
+          end
         end
-        
         @db[@table_name.to_sym].multi_insert(data, return: :primary_key)
       else
         upsert_texts(texts: texts, ids: ids)

data/lib/bx_builder_chain/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module BxBuilderChain
-  VERSION = "0.1.12"
+  VERSION = "0.1.13"
 end

data/lib/bx_builder_chain.rb

@@ -3,6 +3,7 @@
 require "logger"
 require 'dotenv'
 require "zeitwerk"
+require "async"
 require_relative 'bx_builder_chain/configuration'
 
 Dotenv.load

data/lib/generators/bx_builder_chain/templates/app/services/bx_builder_chain/question_asking_service.rb

@@ -2,12 +2,13 @@ module BxBuilderChain
     class QuestionAskingService
       attr_reader :question, :user_groups, :llm_class_name, :client_class_name, :context_results
 
-      def initialize(question:, user_groups: ['public'], client_class_name: 'BxBuilderChain::Vectorsearch::Pgvector', llm_class_name: 'BxBuilderChain::Llm::OpenAi', context_results: 6)
+      def initialize(question:, user_groups: ['public'], client_class_name: 'BxBuilderChain::Vectorsearch::Pgvector', llm_class_name: 'BxBuilderChain::Llm::OpenAi', context_results: 6, model: "gpt-3.5-turbo-16k")
         @question = question
         @user_groups = user_groups
         @client_class_name = client_class_name
         @llm_class_name = llm_class_name
         @context_results = context_results
+        @model = model
       end
   
       def ask
@@ -16,20 +17,54 @@ module BxBuilderChain
         response = client.ask(question: question, context_results: context_results)
         { answer: response }
       end
+
+      # history in the format of
+      # [{role:'human', message:'user message', {role:'ai',message:'some text'}, .....]
+      def chat(history = [])
+        search_results = client.similarity_search(query: @question, k: @context_results)
+
+        context = search_results.map do |result|
+          result.content.to_s
+        end
+        context = context.join("\n---\n")
   
-      private
+        prompt = client.generate_prompt(question: @question, context: context, prompt_template: nil)
   
+        llm.chat(prompt:prompt, messages: reduce_history(history))
+      end
+
+      def reduce_history(history)
+        current_length = 0
+        limit = tokenizer.class::TOKEN_LIMITS["gpt-3.5-turbo-16k"] * 0.75
+        history_overflow, @reduced_history = @history.partition do |msg|
+          current_length += tokenizer.token_length(msg[:message], @current_model)
+          current_length > limit
+        end
+
+        reduced_history
+      end
+
+      private
+
+      def tokenizer
+        @tokenizer ||= llm.class::LENGTH_VALIDATOR
+      end
+      
       def client
         @client ||= client_class_name.constantize.new(
-          llm: llm_class_name.constantize.new(
-            default_options: {
-              chat_completion_model_name: "gpt-3.5-turbo", 
-              temperature: 0.2
-            }
-          ),
+          llm: llm,
           namespaces: user_groups
         )
       end
+
+      def llm
+        @llm ||= llm_class_name.constantize.new(
+          default_options: {
+            chat_completion_model_name: "gpt-3.5-turbo-16k", 
+            temperature: 0.2
+          }
+        )
+      end
     end
   end
   

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bx_builder_chain
 version: !ruby/object:Gem::Version
-  version: 0.1.12
+  version: 0.1.13
 platform: ruby
 authors:
 - Paul Ketelle
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-09-29 00:00:00.000000000 Z
+date: 2024-01-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: zeitwerk
@@ -150,6 +150,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: 2.8.3
+- !ruby/object:Gem::Dependency
+  name: async
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.31.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.31.0
 description: LangChain but for the builder tech stack.
 email:
 - paul.ketelle@builder.ai