langchainrb 0.6.9 → 0.6.11

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4d046a7cae545da3092694fee60b0fa3caa0852492c9bf7dbbe35001d96106e4
-  data.tar.gz: 8ed841d2359400f9a4700a49b82030fbe3dd3563545eca31b487914e8fb236f7
+  metadata.gz: 78d1726db950b4234799cdf34185110411cbc666836225a63ed9d5be6b0a5575
+  data.tar.gz: 0ed5595b0aae9dcaa97ee5cbc653cf3c9cb21f6057f5439ec78711ecb1cf6b46
 SHA512:
-  metadata.gz: eed2f03cbdd74ef42fafd07524dc6bf5c7ac2278e5e55dd1aad5b2cd87f319a91c8719a6e9ac43ac50cda5525f3defab14cfecc749bdec81075d52b3935b23dd
-  data.tar.gz: cd4fce88b0f9f545f6d32660879e6fa5044ec74b41dcb41bc29eafb78b307d8b2cb8204856098657d8c9707331c8629c8a707a227fe4d20a5816347f8a66991c
+  metadata.gz: c0313618b12b10943e8825aaabd66306e21b0a24fefe9c4593cb4f28b1a03e099781944057ce9423c98851bad7dea07e2832c4ab37d416cf743b126a26a6a308
+  data.tar.gz: a2910e93f883bee84288e5bd22644f17b30957a6ae9a696b3adc01dc9b32e2f546b49d46c3587856a883cd4dc1f0a677a3970d357b5fe2dfd9216ce894d06120

data/.env.example

@@ -1,4 +1,5 @@
 AI21_API_KEY=
+ANTHROPIC_API_KEY=
 CHROMA_URL=
 COHERE_API_KEY=
 GOOGLE_PALM_API_KEY=

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 ## [Unreleased]
 
+## [0.6.10] - 2023-08-01
+- 🗣️ LLMs
+  - Introducing Anthropic support
+
 ## [0.6.9] - 2023-07-29
 
 ## [0.6.8] - 2023-07-21

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    langchainrb (0.6.9)
+    langchainrb (0.6.11)
       baran (~> 0.1.6)
       colorize (~> 0.8.1)
       json-schema (~> 4.0.0)
@@ -33,6 +33,9 @@ GEM
       public_suffix (>= 2.0.2, < 6.0)
     afm (0.2.2)
     ai21 (0.2.1)
+    anthropic (0.1.0)
+      faraday (>= 1)
+      faraday-multipart (>= 1)
     ast (2.4.2)
     baran (0.1.7)
     builder (3.2.4)
@@ -223,7 +226,7 @@ GEM
       zeitwerk (~> 2.5)
     rainbow (3.1.1)
     rake (13.0.6)
-    rb_sys (0.9.79)
+    rb_sys (0.9.81)
     rdiscount (2.2.7)
     regexp_parser (2.8.0)
     replicate-ruby (0.2.2)
@@ -318,6 +321,7 @@ PLATFORMS
 
 DEPENDENCIES
   ai21 (~> 0.2.1)
+  anthropic (~> 0.1.0)
   chroma-db (~> 0.3.0)
   cohere-ruby (~> 0.9.5)
   docx (~> 0.8.0)

data/README.md

@@ -203,6 +203,16 @@ Add `gem "ai21", "~> 0.2.1"` to your Gemfile.
 ai21 = Langchain::LLM::AI21.new(api_key: ENV["AI21_API_KEY"])
 ```
 
+#### Anthropic
+Add `gem "anthropic", "~> 0.1.0"` to your Gemfile.
+```ruby
+anthropic = Langchain::LLM::Anthropic.new(api_key: ENV["ANTHROPIC_API_KEY"])
+```
+
+```ruby
+anthropic.complete(prompt: "What is the meaning of life?")
+```
+
 ### Using Prompts 📋
 
 #### Prompt Templates
@@ -537,7 +547,7 @@ Join us in the [Langchain.rb](https://discord.gg/WDARp7J2n8) Discord server.
 
 ## Contributing
 
-Bug reports and pull requests are welcome on GitHub at https://github.com/andreibondarev/langchain.
+Bug reports and pull requests are welcome on GitHub at https://github.com/andreibondarev/langchainrb.
 
 ## License
 

data/examples/open_ai_qdrant_function_calls.rb

@@ -11,10 +11,6 @@ functions = [
         controller_name: {
           type: :string,
           description: "the controller name, e.g. users_controller"
-        },
-        unit: {
-          type: "string",
-          enum: %w[celsius fahrenheit]
         }
       },
       required: ["controller_name"]

data/lib/langchain/llm/anthropic.rb

@@ -0,0 +1,63 @@
+# frozen_string_literal: true
+
+module Langchain::LLM
+  #
+  # Wrapper around Anthropic APIs.
+  #
+  # Gem requirements:
+  #   gem "anthropic", "~> 0.1.0"
+  #
+  # Usage:
+  #     anthorpic = Langchain::LLM::Anthropic.new(api_key:)
+  #
+  class Anthropic < Base
+    DEFAULTS = {
+      temperature: 0.0,
+      completion_model_name: "claude-2",
+      max_tokens_to_sample: 256
+    }.freeze
+
+    # TODO: Implement token length validator for Anthropic
+    # LENGTH_VALIDATOR = Langchain::Utils::TokenLength::AnthropicValidator
+
+    def initialize(api_key:, llm_options: {}, default_options: {})
+      depends_on "anthropic"
+      require "anthropic"
+
+      @client = ::Anthropic::Client.new(access_token: api_key, **llm_options)
+      @defaults = DEFAULTS.merge(default_options)
+    end
+
+    #
+    # Generate a completion for a given prompt
+    #
+    # @param prompt [String] The prompt to generate a completion for
+    # @param params [Hash] extra parameters passed to Anthropic::Client#complete
+    # @return [String] The completion
+    #
+    def complete(prompt:, **params)
+      parameters = compose_parameters @defaults[:completion_model_name], params
+
+      parameters[:prompt] = prompt
+
+      # TODO: Implement token length validator for Anthropic
+      # parameters[:max_tokens_to_sample] = validate_max_tokens(prompt, parameters[:completion_model_name])
+
+      response = client.complete(parameters: parameters)
+      response.dig("completion")
+    end
+
+    private
+
+    def compose_parameters(model, params)
+      default_params = {model: model}.merge(@defaults.except(:completion_model_name))
+
+      default_params.merge(params)
+    end
+
+    # TODO: Implement token length validator for Anthropic
+    # def validate_max_tokens(messages, model)
+    #   LENGTH_VALIDATOR.validate_max_tokens!(messages, model)
+    # end
+  end
+end

data/lib/langchain/loader.rb

@@ -89,9 +89,9 @@ module Langchain
     end
 
     def load_from_path
-      raise FileNotFound unless File.exist?(@path)
+      return File.open(@path) if File.exist?(@path)
 
-      File.open(@path)
+      raise FileNotFound, "File #{@path} does not exist"
     end
 
     def load_from_directory(&block)

data/lib/langchain/vectorsearch/milvus.rb

@@ -8,7 +8,7 @@ module Langchain::Vectorsearch
     # Gem requirements: gem "milvus", "~> 0.9.0"
     #
     # Usage:
-    # milvus = Langchain::Vectorsearch::Milvus.new(url:, index_name:, llm:, llm_api_key:)
+    # milvus = Langchain::Vectorsearch::Milvus.new(url:, index_name:, llm:, api_key:)
     #
 
     def initialize(url:, index_name:, llm:, api_key: nil)
@@ -32,7 +32,7 @@ module Langchain::Vectorsearch
             field: Array(texts)
           }, {
             field_name: "vectors",
-            type: ::Milvus::DATA_TYPES["binary_vector"],
+            type: ::Milvus::DATA_TYPES["float_vector"],
             field: Array(texts).map { |text| llm.embed(text: text) }
           }
         ]
@@ -47,7 +47,7 @@ module Langchain::Vectorsearch
       client.collections.create(
         auto_id: true,
         collection_name: index_name,
-        description: "Default schema created by Vectorsearch",
+        description: "Default schema created by langchain.rb",
         fields: [
           {
             name: "id",
@@ -66,7 +66,7 @@ module Langchain::Vectorsearch
             ]
           }, {
             name: "vectors",
-            data_type: ::Milvus::DATA_TYPES["binary_vector"],
+            data_type: ::Milvus::DATA_TYPES["float_vector"],
             is_primary_key: false,
             type_params: [
               {
@@ -79,6 +79,20 @@ module Langchain::Vectorsearch
       )
     end
 
+    # Create the default index
+    # @return [Boolean] The response from the server
+    def create_default_index
+      client.indices.create(
+        collection_name: "Documents",
+        field_name: "vectors",
+        extra_params: [
+          {key: "metric_type", value: "L2"},
+          {key: "index_type", value: "IVF_FLAT"},
+          {key: "params", value: "{\"nlist\":1024}"}
+        ]
+      )
+    end
+
     # Get the default schema
     # @return [Hash] The response from the server
     def get_default_schema
@@ -91,6 +105,12 @@ module Langchain::Vectorsearch
       client.collections.delete(collection_name: index_name)
     end
 
+    # Load default schema into memory
+    # @return [Boolean] The response from the server
+    def load_default_schema
+      client.collections.load(collection_name: index_name)
+    end
+
     def similarity_search(query:, k: 4)
       embedding = llm.embed(text: query)
 
@@ -101,15 +121,36 @@ module Langchain::Vectorsearch
     end
 
     def similarity_search_by_vector(embedding:, k: 4)
+      load_default_schema
+
       client.search(
         collection_name: index_name,
+        output_fields: ["id", "content", "vectors"],
         top_k: k.to_s,
         vectors: [embedding],
         dsl_type: 1,
         params: "{\"nprobe\": 10}",
-        anns_field: "content",
-        metric_type: "L2"
+        anns_field: "vectors",
+        metric_type: "L2",
+        vector_type: ::Milvus::DATA_TYPES["float_vector"]
       )
     end
+
+    # Ask a question and return the answer
+    # @param question [String] The question to ask
+    # @yield [String] Stream responses back one String at a time
+    # @return [String] The answer to the question
+    def ask(question:, &block)
+      search_results = similarity_search(query: question)
+
+      content_field = search_results.dig("results", "fields_data").select { |field| field.dig("field_name") == "content" }
+      content_data = content_field.first.dig("Field", "Scalars", "Data", "StringData", "data")
+
+      context = content_data.join("\n---\n")
+
+      prompt = generate_prompt(question: question, context: context)
+
+      llm.chat(prompt: prompt, &block)
+    end
   end
 end

data/lib/langchain/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Langchain
-  VERSION = "0.6.9"
+  VERSION = "0.6.11"
 end

data/lib/langchain.rb

@@ -131,6 +131,7 @@ module Langchain
 
   module LLM
     autoload :AI21, "langchain/llm/ai21"
+    autoload :Anthropic, "langchain/llm/anthropic"
     autoload :Base, "langchain/llm/base"
     autoload :Cohere, "langchain/llm/cohere"
     autoload :GooglePalm, "langchain/llm/google_palm"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: langchainrb
 version: !ruby/object:Gem::Version
-  version: 0.6.9
+  version: 0.6.11
 platform: ruby
 authors:
 - Andrei Bondarev
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-07-29 00:00:00.000000000 Z
+date: 2023-08-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: baran
@@ -136,6 +136,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: 0.2.1
+- !ruby/object:Gem::Dependency
+  name: anthropic
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 0.1.0
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 0.1.0
 - !ruby/object:Gem::Dependency
   name: chroma-db
   requirement: !ruby/object:Gem::Requirement
@@ -518,6 +532,7 @@ files:
 - lib/langchain/data.rb
 - lib/langchain/dependency_helper.rb
 - lib/langchain/llm/ai21.rb
+- lib/langchain/llm/anthropic.rb
 - lib/langchain/llm/base.rb
 - lib/langchain/llm/cohere.rb
 - lib/langchain/llm/google_palm.rb