boxcars 0.4.9 → 0.4.10

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b1199d0161dfab186b1fd3e85789a86d242d5351a0439bbdec719d0d55eecdaa
-  data.tar.gz: 4f0c4b90d023c7a9bb86d45b0aa01ff0d7805bdf7d5f2b36401c87525eb15403
+  metadata.gz: 4e9d011a30aa2b478281dca050fec260a09f5e985454beee40405d0ae6991411
+  data.tar.gz: 2bd6e365273101ebd40a35780a1ef9913f4125a29dc24acaab317146a7c19344
 SHA512:
-  metadata.gz: abc63d10170ba704a0cc1bffc60dc2cc29ffa6927e945050d7157a7779d576be25bf8ed3c738fa6be0ec1e7fdc898ba2d41fc807026e7a2a1232ce7f68d3f283
-  data.tar.gz: 85cf33b3832230a9294557f1a23f305c97a1c8bc31414849160118524a64d1e27bdd74d7b1e186895596b9ce9e82afd6b8fb1fe9e7b081b0e33617228a947834
+  metadata.gz: f8937c0f246c0488e0656646bdcd9a02a62edbb068fbd92710e2e275c99c8c7006e4bfec277206b63ae5507833481ec2db9b0a274bb74be743d4739e8bfacc98
+  data.tar.gz: 5c176f8854d058cad2426786707c11a49759781167e1a861b1e376da1f7c9b679e75b7a33645134cc3e8d7298e907c0c831059328979ff8d590262b5804e76d8

data/Gemfile

@@ -7,7 +7,7 @@ gemspec
 
 gem "debug", "~> 1.9"
 
-gem "dotenv", "~> 2.8"
+gem "dotenv", "~> 3.1"
 
 gem "rake", "~> 13.1"
 
@@ -31,10 +31,10 @@ gem "pg", "~> 1.5"
 gem "pgvector", "~> 0.2.2"
 
 group :development, :test do
-  gem "rspec", "~> 3.2"
+  gem "rspec", "~> 3.13"
   gem "rubocop", "~> 1.60"
   gem "vcr", "~> 6.2.0"
-  gem "webmock", "~> 3.19.1"
+  gem "webmock", "~> 3.23.0"
   gem "rubocop-rake", "~> 0.6.0"
-  gem "rubocop-rspec", "~> 2.26"
+  gem "rubocop-rspec", "~> 2.29"
 end

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    boxcars (0.4.9)
+    boxcars (0.4.10)
       anthropic (~> 0.1)
       google_search_results (~> 2.2)
       gpt4all (~> 0.0.4)
@@ -48,31 +48,31 @@ GEM
       async
     async-pool (0.4.0)
       async (>= 1.25)
-    base64 (0.2.0)
+    bigdecimal (3.1.6)
     concurrent-ruby (1.2.2)
     console (1.17.4)
       fiber-annotation
       fiber-local
-    crack (0.4.5)
+    crack (1.0.0)
+      bigdecimal
       rexml
     debug (1.9.1)
       irb (~> 1.10)
       reline (>= 0.3.8)
-    diff-lcs (1.5.0)
+    diff-lcs (1.5.1)
     domain_name (0.5.20190701)
       unf (>= 0.0.5, < 1.0.0)
-    dotenv (2.8.1)
+    dotenv (3.1.0)
     event_stream_parser (1.0.0)
-    faraday (2.7.12)
-      base64
-      faraday-net_http (>= 2.0, < 3.1)
-      ruby2_keywords (>= 0.0.4)
+    faraday (2.9.0)
+      faraday-net_http (>= 2.0, < 3.2)
     faraday-http-cache (2.5.0)
       faraday (>= 0.8)
     faraday-multipart (1.0.4)
       multipart-post (~> 2)
-    faraday-net_http (3.0.2)
-    faraday-retry (2.2.0)
+    faraday-net_http (3.1.0)
+      net-http
+    faraday-retry (2.2.1)
       faraday (~> 2.0)
     fiber-annotation (0.2.0)
     fiber-local (1.0.0)
@@ -101,7 +101,7 @@ GEM
     irb (1.11.0)
       rdoc
       reline (>= 0.3.8)
-    json (2.7.1)
+    json (2.7.2)
     language_server-protocol (3.17.0.3)
     mime-types (3.4.1)
       mime-types-data (~> 3.2015)
@@ -109,11 +109,13 @@ GEM
     minitest (5.20.0)
     multi_json (1.15.0)
     multipart-post (2.3.0)
+    net-http (0.4.1)
+      uri
     netrc (0.11.0)
     nio4r (2.5.9)
-    nokogiri (1.16.0-arm64-darwin)
+    nokogiri (1.16.2-arm64-darwin)
       racc (~> 1.4)
-    nokogiri (1.16.0-x86_64-linux)
+    nokogiri (1.16.2-x86_64-linux)
       racc (~> 1.4)
     octokit (4.25.1)
       faraday (>= 1, < 3)
@@ -123,7 +125,7 @@ GEM
     parser (3.3.0.5)
       ast (~> 2.4.1)
       racc
-    pg (1.5.4)
+    pg (1.5.6)
     pgvector (0.2.2)
     protocol-hpack (1.4.2)
     protocol-http (0.25.0)
@@ -138,7 +140,7 @@ GEM
     racc (1.7.3)
     rainbow (3.1.1)
     rake (13.1.0)
-    rdoc (6.6.2)
+    rdoc (6.6.3.1)
       psych (>= 4.0.0)
     regexp_parser (2.9.0)
     reline (0.4.1)
@@ -149,19 +151,19 @@ GEM
       mime-types (>= 1.16, < 4.0)
       netrc (~> 0.8)
     rexml (3.2.6)
-    rspec (3.12.0)
-      rspec-core (~> 3.12.0)
-      rspec-expectations (~> 3.12.0)
-      rspec-mocks (~> 3.12.0)
-    rspec-core (3.12.2)
-      rspec-support (~> 3.12.0)
-    rspec-expectations (3.12.3)
+    rspec (3.13.0)
+      rspec-core (~> 3.13.0)
+      rspec-expectations (~> 3.13.0)
+      rspec-mocks (~> 3.13.0)
+    rspec-core (3.13.0)
+      rspec-support (~> 3.13.0)
+    rspec-expectations (3.13.0)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.12.0)
-    rspec-mocks (3.12.6)
+      rspec-support (~> 3.13.0)
+    rspec-mocks (3.13.0)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.12.0)
-    rspec-support (3.12.1)
+      rspec-support (~> 3.13.0)
+    rspec-support (3.13.1)
     rubocop (1.60.2)
       json (~> 2.3)
       language_server-protocol (>= 3.17.0)
@@ -173,29 +175,31 @@ GEM
       rubocop-ast (>= 1.30.0, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 3.0)
-    rubocop-ast (1.30.0)
-      parser (>= 3.2.1.0)
+    rubocop-ast (1.31.2)
+      parser (>= 3.3.0.4)
     rubocop-capybara (2.20.0)
       rubocop (~> 1.41)
     rubocop-factory_bot (2.25.1)
       rubocop (~> 1.41)
     rubocop-rake (0.6.0)
       rubocop (~> 1.0)
-    rubocop-rspec (2.26.1)
+    rubocop-rspec (2.29.1)
       rubocop (~> 1.40)
       rubocop-capybara (~> 2.17)
       rubocop-factory_bot (~> 2.22)
+      rubocop-rspec_rails (~> 2.28)
+    rubocop-rspec_rails (2.28.2)
+      rubocop (~> 1.40)
     ruby-openai (6.3.1)
       event_stream_parser (>= 0.3.0, < 2.0.0)
       faraday (>= 1)
       faraday-multipart (>= 1)
     ruby-progressbar (1.13.0)
-    ruby2_keywords (0.0.5)
     sawyer (0.9.2)
       addressable (>= 2.3.5)
       faraday (>= 0.17.3, < 3)
-    sqlite3 (1.7.1-arm64-darwin)
-    sqlite3 (1.7.1-x86_64-linux)
+    sqlite3 (1.7.2-arm64-darwin)
+    sqlite3 (1.7.2-x86_64-linux)
     stringio (3.1.0)
     strings-ansi (0.2.0)
     timers (4.3.5)
@@ -213,8 +217,9 @@ GEM
       unf_ext
     unf_ext (0.0.8.2)
     unicode-display_width (2.5.0)
+    uri (0.13.0)
     vcr (6.2.0)
-    webmock (3.19.1)
+    webmock (3.23.0)
       addressable (>= 2.8.0)
       crack (>= 0.3.2)
       hashdiff (>= 0.4.0, < 2.0.0)
@@ -230,7 +235,7 @@ DEPENDENCIES
   async (~> 1.31.0)
   boxcars!
   debug (~> 1.9)
-  dotenv (~> 2.8)
+  dotenv (~> 3.1)
   faraday-retry (~> 2.0)
   github_changelog_generator (~> 1.16)
   hnswlib (~> 0.9.0)
@@ -238,13 +243,13 @@ DEPENDENCIES
   pgvector (~> 0.2.2)
   rake (~> 13.1)
   rest-client (~> 2.1)
-  rspec (~> 3.2)
+  rspec (~> 3.13)
   rubocop (~> 1.60)
   rubocop-rake (~> 0.6.0)
-  rubocop-rspec (~> 2.26)
+  rubocop-rspec (~> 2.29)
   sqlite3 (~> 1.7)
   vcr (~> 6.2.0)
-  webmock (~> 3.19.1)
+  webmock (~> 3.23.0)
 
 BUNDLED WITH
    2.4.16

data/README.md

@@ -46,7 +46,7 @@ Or install it yourself as:
 
 ## Usage
 
-We will be adding more examples soon, but here are a couple to get you started. First, you'll need to set up your environment variables for OpenAI and Google SERP (OPENAI_ACCESS_TOKEN, SERPAPI_API_KEY). If you prefer not to set these variables in your environment, you can pass them directly into the API.
+We will be adding more examples soon, but here are a couple to get you started. First, you'll need to set up your environment variables for services like OpenAI, Anthropic, and Google SERP (OPENAI_ACCESS_TOKEN, ANTHROPIC_API_KEY,SERPAPI_API_KEY) etc. If you prefer not to set these variables in your environment, you can pass them directly into the API.
 
 In the examples below, we added one Ruby gem to load the environment at the first line, but depending on what you want, you might not need this.
 ```ruby

data/lib/boxcars/boxcar/google_search.rb

@@ -25,8 +25,8 @@ module Boxcars
     def run(question)
       search = ::GoogleSearch.new(q: question)
       rv = find_answer(search.get_hash)
-      puts "Question: #{question}"
-      puts "Answer: #{rv}"
+      Boxcars.info "Question: #{question}"
+      Boxcars.info "Answer: #{rv}"
       rv
     end
 

data/lib/boxcars/engine/cohere.rb

@@ -0,0 +1,187 @@
+# frozen_string_literal: true
+
+# Boxcars - a framework for running a series of tools to get an answer to a question.
+module Boxcars
+  # A engine that uses Cohere's API.
+  class Cohere < Engine
+    attr_reader :prompts, :llm_params, :model_kwargs, :batch_size
+
+    # The default parameters to use when asking the engine.
+    DEFAULT_PARAMS = {
+      model: "command-r-plus",
+      max_tokens: 4000,
+      max_input_tokens: 1000,
+      temperature: 0.2
+    }.freeze
+
+    # the default name of the engine
+    DEFAULT_NAME = "Cohere engine"
+    # the default description of the engine
+    DEFAULT_DESCRIPTION = "useful for when you need to use Cohere AI to answer questions. " \
+                          "You should ask targeted questions"
+
+    # A engine is the driver for a single tool to run.
+    # @param name [String] The name of the engine. Defaults to "OpenAI engine".
+    # @param description [String] A description of the engine. Defaults to:
+    #        useful for when you need to use AI to answer questions. You should ask targeted questions".
+    # @param prompts [Array<String>] The prompts to use when asking the engine. Defaults to [].
+    def initialize(name: DEFAULT_NAME, description: DEFAULT_DESCRIPTION, prompts: [], **kwargs)
+      @llm_params = DEFAULT_PARAMS.merge(kwargs)
+      @prompts = prompts
+      @batch_size = 20
+      super(description: description, name: name)
+    end
+
+    def conversation_model?(_model)
+      true
+    end
+
+    def chat(params, cohere_api_key)
+      raise Boxcars::ConfigurationError('Cohere API key not set') if cohere_api_key.blank?
+
+      # Define the API endpoint and parameters
+      api_endpoint = 'https://api.cohere.ai/v1/chat'
+
+      connection = Faraday.new(api_endpoint) do |faraday|
+        faraday.request :url_encoded
+        faraday.headers['Authorization'] = "Bearer #{cohere_api_key}"
+        faraday.headers['Content-Type'] = 'application/json'
+      end
+
+      # Make the API call
+      response = connection.post { |req| req.body = params.to_json }
+
+      # response_data = JSON.parse(response.body, symbolize_names: true)
+      # response_data[:text]
+      JSON.parse(response.body, symbolize_names: true)
+    end
+
+    # Get an answer from the engine.
+    # @param prompt [String] The prompt to use when asking the engine.
+    # @param cohere_api_key [String] Optional api key to use when asking the engine.
+    #   Defaults to Boxcars.configuration.cohere_api_key.
+    # @param kwargs [Hash] Additional parameters to pass to the engine if wanted.
+    def client(prompt:, inputs: {}, **kwargs)
+      api_key = Boxcars.configuration.cohere_api_key(**kwargs)
+      params = prompt.as_prompt(inputs: inputs, prefixes: default_prefixes, show_roles: true).merge(llm_params.merge(kwargs))
+      params[:message] = params.delete(:prompt)
+      params[:stop_sequences] = params.delete(:stop) if params.key?(:stop)
+      Boxcars.debug("Prompt after formatting:#{params[:message]}", :cyan) if Boxcars.configuration.log_prompts
+      chat(params, api_key)
+    end
+
+    # get an answer from the engine for a question.
+    # @param question [String] The question to ask the engine.
+    # @param kwargs [Hash] Additional parameters to pass to the engine if wanted.
+    def run(question, **kwargs)
+      prompt = Prompt.new(template: question)
+      response = client(prompt: prompt, **kwargs)
+
+      raise Error, "Cohere: No response from API" unless response
+      raise Error, "Cohere: #{response[:error]}" if response[:error]
+
+      answer = response[:text]
+      Boxcars.debug(response, :yellow)
+      answer
+    end
+
+    # Get the default parameters for the engine.
+    def default_params
+      llm_params
+    end
+
+    # Get generation informaton
+    # @param sub_choices [Array<Hash>] The choices to get generation info for.
+    # @return [Array<Generation>] The generation information.
+    def generation_info(sub_choices)
+      sub_choices.map do |choice|
+        Generation.new(
+          text: choice["completion"],
+          generation_info: {
+            finish_reason: choice.fetch("stop_reason", nil),
+            logprobs: choice.fetch("logprobs", nil)
+          }
+        )
+      end
+    end
+
+    # make sure we got a valid response
+    # @param response [Hash] The response to check.
+    # @param must_haves [Array<String>] The keys that must be in the response. Defaults to %w[choices].
+    # @raise [KeyError] if there is an issue with the access token.
+    # @raise [ValueError] if the response is not valid.
+    def check_response(response, must_haves: %w[completion])
+      if response['error']
+        code = response.dig('error', 'code')
+        msg = response.dig('error', 'message') || 'unknown error'
+        raise KeyError, "ANTHOPIC_API_KEY not valid" if code == 'invalid_api_key'
+
+        raise ValueError, "Cohere error: #{msg}"
+      end
+
+      must_haves.each do |key|
+        raise ValueError, "Expecting key #{key} in response" unless response.key?(key)
+      end
+    end
+
+    # Call out to OpenAI's endpoint with k unique prompts.
+    # @param prompts [Array<String>] The prompts to pass into the model.
+    # @param inputs [Array<String>] The inputs to subsitite into the prompt.
+    # @param stop [Array<String>] Optional list of stop words to use when generating.
+    # @return [EngineResult] The full engine output.
+    def generate(prompts:, stop: nil)
+      params = {}
+      params[:stop] = stop if stop
+      choices = []
+      # Get the token usage from the response.
+      # Includes prompt, completion, and total tokens used.
+      prompts.each_slice(batch_size) do |sub_prompts|
+        sub_prompts.each do |sprompts, inputs|
+          response = client(prompt: sprompts, inputs: inputs, **params)
+          check_response(response)
+          choices << response
+        end
+      end
+
+      n = params.fetch(:n, 1)
+      generations = []
+      prompts.each_with_index do |_prompt, i|
+        sub_choices = choices[i * n, (i + 1) * n]
+        generations.push(generation_info(sub_choices))
+      end
+      EngineResult.new(generations: generations, engine_output: { token_usage: {} })
+    end
+    # rubocop:enable Metrics/AbcSize
+
+    # the engine type
+    def engine_type
+      "claude"
+    end
+
+    # calculate the number of tokens used
+    def get_num_tokens(text:)
+      text.split.length # TODO: hook up to token counting gem
+    end
+
+    # lookup the context size for a model by name
+    # @param modelname [String] The name of the model to lookup.
+    def modelname_to_contextsize(_modelname)
+      100000
+    end
+
+    # Calculate the maximum number of tokens possible to generate for a prompt.
+    # @param prompt_text [String] The prompt text to use.
+    # @return [Integer] the number of tokens possible to generate.
+    def max_tokens_for_prompt(prompt_text)
+      num_tokens = get_num_tokens(prompt_text)
+
+      # get max context size for model by name
+      max_size = modelname_to_contextsize(model_name)
+      max_size - num_tokens
+    end
+
+    def default_prefixes
+      { system: "SYSTEM: ", user: "USER: ", assistant: "CHATBOT: ", history: :history }
+    end
+  end
+end

data/lib/boxcars/engine.rb

@@ -21,6 +21,7 @@ end
 
 require "boxcars/engine/engine_result"
 require "boxcars/engine/anthropic"
+require "boxcars/engine/cohere"
 require "boxcars/engine/openai"
 require "boxcars/engine/perplexityai"
 require "boxcars/engine/gpt4all_eng"

data/lib/boxcars/version.rb

@@ -2,5 +2,5 @@
 
 module Boxcars
   # The current version of the gem.
-  VERSION = "0.4.9"
+  VERSION = "0.4.10"
 end

data/lib/boxcars.rb

@@ -52,6 +52,11 @@ module Boxcars
       key_lookup(:anthropic_api_key, kwargs)
     end
 
+    # @return [String] The Anthropic API key either from arg or env.
+    def cohere_api_key(**kwargs)
+      key_lookup(:cohere_api_key, kwargs)
+    end
+
     private
 
     def check_key(key, val)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: boxcars
 version: !ruby/object:Gem::Version
-  version: 0.4.9
+  version: 0.4.10
 platform: ruby
 authors:
 - Francis Sullivan
@@ -9,7 +9,7 @@ authors:
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-01-25 00:00:00.000000000 Z
+date: 2024-04-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: anthropic
@@ -157,6 +157,7 @@ files:
 - lib/boxcars/conversation_prompt.rb
 - lib/boxcars/engine.rb
 - lib/boxcars/engine/anthropic.rb
+- lib/boxcars/engine/cohere.rb
 - lib/boxcars/engine/engine_result.rb
 - lib/boxcars/engine/gpt4all_eng.rb
 - lib/boxcars/engine/openai.rb