active_genie 0.0.12 → 0.0.18

data/lib/active_genie/ranking/ranking_scoring.rb

@@ -31,14 +31,14 @@ module ActiveGenie::Ranking
     end
 
     def generate_score(player)
-      score, reasoning = ActiveGenie::Scoring::Basic.call(
+      score, reasoning = ActiveGenie::Scoring.call(
         player.content,
         @criteria,
         @reviewers,
         config: @config
       ).values_at('final_score', 'final_reasoning')
 
-      ActiveGenie::Logger.debug({step: :new_score, player_id: player.id, score:, reasoning: })
+      ActiveGenie::Logger.debug({ code: :new_score, player_id: player.id, score:, reasoning: })
 
       score
     end
@@ -52,7 +52,7 @@ module ActiveGenie::Ranking
         config: @config
       ).values_at('reviewer1', 'reviewer2', 'reviewer3')
       
-      ActiveGenie::Logger.debug({step: :new_reviewers, reviewers: [reviewer1, reviewer2, reviewer3] })
+      ActiveGenie::Logger.debug({ code: :new_reviewers, reviewers: [reviewer1, reviewer2, reviewer3] })
 
       [reviewer1, reviewer2, reviewer3]
     end

data/lib/active_genie/scoring/basic.rb

@@ -43,48 +43,67 @@ module ActiveGenie::Scoring
         {  role: 'user', content: "Text to score: #{@text}" }, 
       ]
 
-      properties = {}
-      get_or_recommend_reviewers.each do |reviewer|
-        properties["#{reviewer}_reasoning"] = {
-          type: 'string',
-          description: "The reasoning of the scoring process by #{reviewer}.",
-        }
-        properties["#{reviewer}_score"] = {
-          type: 'number',
-          description: "The score given by #{reviewer}.",
-          min: 0,
-          max: 100
-        }
-      end
+      properties = build_properties
 
       function = {
         name: 'scoring',
         description: 'Score the text based on the given criteria.',
         schema: {
           type: "object",
-          properties: {
-            **properties,
-            final_score: {
-              type: 'number',
-              description: 'The final score based on the previous reviewers',
-            },
-            final_reasoning: {
-              type: 'string',
-              description: 'The final reasoning based on the previous reviewers',
-            }
-          }
+          properties:,
+          required: properties.keys
         }
       }
 
-      ::ActiveGenie::Clients::UnifiedClient.function_calling(
+      result = ::ActiveGenie::Clients::UnifiedClient.function_calling(
         messages,
         function,
         model_tier: 'lower_tier',
         config: @config
       )
+
+      result['final_score'] = 0 if result['final_score'].nil?
+
+      ActiveGenie::Logger.debug({
+        code: :scoring,
+        text: @text[0..30],
+        criteria: @criteria[0..30],
+        reviewers: get_or_recommend_reviewers,
+        score: result['final_score'],
+        reasoning: result['final_reasoning']
+      })
+
+      result
     end
 
     private
+    
+    def build_properties
+      properties = {}
+      get_or_recommend_reviewers.each do |reviewer|
+        properties["#{reviewer}_reasoning"] = {
+          type: 'string',
+          description: "The reasoning of the scoring process by #{reviewer}.",
+        }
+        properties["#{reviewer}_score"] = {
+          type: 'number',
+          description: "The score given by #{reviewer}.",
+          min: 0,
+          max: 100
+        }
+      end
+
+      properties[:final_score] = {
+        type: 'number',
+        description: 'The final score based on the previous reviewers',
+      }
+      properties[:final_reasoning] = {
+        type: 'string',
+        description: 'The final reasoning based on the previous reviewers',
+      }
+
+      properties
+    end
 
     def get_or_recommend_reviewers
       @get_or_recommend_reviewers ||= if @reviewers.count > 0 

data/lib/active_genie/scoring.rb

@@ -9,6 +9,9 @@ module ActiveGenie
     def basic(...)
       Basic.call(...)
     end
+    def call(...)
+      Basic.call(...)
+    end
 
     def recommended_reviewers(...)
       RecommendedReviewers.call(...)

data/lib/tasks/benchmark.rake

@@ -0,0 +1,27 @@
+
+namespace :active_genie do
+  desc "Run benchmarks, optionally for a specific module (e.g., rake active_genie:benchmark[data_extractor])"
+  task :benchmark, [:module_name] do |_, args|
+    Rake::TestTask.new(:run_benchmarks) do |t|
+      t.libs << "benchmark"
+      
+      if args[:module_name]
+        module_name = args[:module_name]
+        module_path = "benchmark/test_cases/#{module_name}/"
+        t.test_files = FileList["#{module_path}**/*_test.rb"]
+        puts "Running benchmarks for module: #{module_name}"
+      else
+        t.test_files = FileList["benchmark/test_cases/**/*_test.rb"]
+        puts "Running all benchmarks"
+      end
+      
+      t.warning = false
+    end
+    
+    begin
+      Rake::Task[:run_benchmarks].invoke
+    rescue => e
+      puts e
+    end
+  end
+end

metadata

@@ -1,92 +1,103 @@
 --- !ruby/object:Gem::Specification
 name: active_genie
 version: !ruby/object:Gem::Version
-  version: 0.0.12
+  version: 0.0.18
 platform: ruby
 authors:
 - Radamés Roriz
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2025-03-17 00:00:00.000000000 Z
+date: 2025-04-02 00:00:00.000000000 Z
 dependencies: []
 description: "# ActiveGenie \U0001F9DE‍♂️\n> The lodash for GenAI, stop reinventing
-  the wheel\n\n[![Gem Version](https://badge.fury.io/rb/active_genie.svg?icon=si%3Arubygems)](https://badge.fury.io/rb/active_genie)\n[![Ruby](https://github.com/roriz/active_genie/actions/workflows/ruby.yml/badge.svg)](https://github.com/roriz/active_genie/actions/workflows/ruby.yml)\n\nActiveGenie
-  is a Ruby gem that provides a polished, production-ready interface for working with
-  Generative AI (GenAI) models. Just like Lodash or ActiveStorage, ActiveGenie simplifies
-  GenAI integration in your Ruby applications.\n\n## Features\n\n- \U0001F3AF **Data
-  Extraction**: Extract structured data from unstructured text with type validation\n-
-  \U0001F4CA **Data Scoring**: Multi-reviewer evaluation system\n- ⚔️ **Data Battle**:
-  Battle between two data like a political debate\n- \U0001F4AD **Data Ranking**:
-  Consistent rank data using scoring + elo ranking + battles\n\n## Installation\n\n1.
-  Add to your Gemfile:\n```ruby\ngem 'active_genie'\n```\n\n2. Install the gem:\n```shell\nbundle
-  install\n```\n\n3. Generate the configuration:\n```shell\necho \"ActiveGenie.load_tasks\"
-  >> Rakefile\nrails g active_genie:install\n```\n\n4. Configure your credentials
-  in `config/initializers/active_genie.rb`:\n```ruby\nActiveGenie.configure do |config|\n
-  \ config.openai.api_key = ENV['OPENAI_API_KEY']\nend\n```\n\n## Quick Start\n\n###
-  Data Extractor\nExtract structured data from text using AI-powered analysis, handling
-  informal language and complex expressions.\n\n```ruby\ntext = \"Nike Air Max 90
-  - Size 42 - $199.99\"\nschema = {\n  brand: { \n    type: 'string',\n    enum: [\"Nike\",
-  \"Adidas\", \"Puma\"]\n  },\n  price: { \n    type: 'number',\n    minimum: 0\n
-  \ },\n  size: {\n    type: 'integer',\n    minimum: 35,\n    maximum: 46\n  }\n}\n\nresult
-  = ActiveGenie::DataExtractor.call(text, schema)\n# => { \n#      brand: \"Nike\",
-  \n#      brand_explanation: \"Brand name found at start of text\",\n#      price:
-  199.99,\n#      price_explanation: \"Price found in USD format at end\",\n#      size:
-  42,\n#      size_explanation: \"Size explicitly stated in the middle\"\n#    }\n```\n\nFeatures:\n-
-  Structured data extraction with type validation\n- Schema-based extraction with
-  custom constraints\n- Informal text analysis (litotes, hedging)\n- Detailed explanations
-  for extracted values\n\nSee the [Data Extractor README](lib/active_genie/data_extractor/README.md)
+  the wheel\n\n[![Gem Version](https://badge.fury.io/rb/active_genie.svg?icon=si%3Arubygems)](https://badge.fury.io/rb/active_genie)\n[![Ruby](https://github.com/roriz/active_genie/actions/workflows/benchmark.yml/badge.svg)](https://github.com/roriz/active_genie/actions/workflows/benchmark.yml)\n\nActiveGenie
+  is a Ruby gem that provides valuable solutions powered by Generative AI (GenAI)
+  models. Just like Lodash or ActiveStorage, ActiveGenie brings a set of Modules reach
+  real value fast and reliable.\nActiveGenie is backed by a custom benchmarking system
+  that ensures consistent quality and performance across different models and providers
+  in every release.\n\n## Installation\n\n1. Add to your Gemfile:\n```ruby\ngem 'active_genie'\n```\n\n2.
+  Install the gem:\n```shell\nbundle install\n```\n\n3. Generate the configuration:\n```shell\necho
+  \"ActiveGenie.load_tasks\" >> Rakefile\nrails g active_genie:install\n```\n\n4.
+  Configure your credentials in `config/initializers/active_genie.rb`:\n```ruby\nActiveGenie.configure
+  do |config|\n  config.openai.api_key = ENV['OPENAI_API_KEY']\nend\n```\n\n## Quick
+  Start\n\n### Data Extractor\n\nExtract structured data from text using AI-powered
+  analysis, handling informal language and complex expressions.\n\n```ruby\ntext =
+  \"Nike Air Max 90 - Size 42 - $199.99\"\nschema = {\n  brand: { \n    type: 'string',\n
+  \   enum: [\"Nike\", \"Adidas\", \"Puma\"]\n  },\n  price: { \n    type: 'number',\n
+  \   minimum: 0\n  },\n  size: {\n    type: 'number',\n    minimum: 35,\n    maximum:
+  46\n  }\n}\n\nresult = ActiveGenie::DataExtractor.call(\n  text,\n  schema,\n  config:
+  { provider: :openai, model: 'gpt-4o-mini' } # optional\n)\n# => { \n#      brand:
+  \"Nike\", \n#      brand_explanation: \"Brand name found at start of text\",\n#
+  \     price: 199.99,\n#      price_explanation: \"Price found in USD format at end\",\n#
+  \     size: 42,\n#      size_explanation: \"Size explicitly stated in the middle\"\n#
+  \   }\n```\n\n*Recommended model*: `gpt-4o-mini`\n\nFeatures:\n- Structured data
+  extraction with type validation\n- Schema-based extraction with custom constraints\n-
+  Informal text analysis (litotes, hedging)\n- Detailed explanations for extracted
+  values\n\nSee the [Data Extractor README](lib/active_genie/data_extractor/README.md)
   for informal text processing, advanced schemas, and detailed interface documentation.\n\n###
-  Data Scoring\nText evaluation system that provides detailed scoring and feedback
-  using multiple expert reviewers. Get balanced scoring through AI-powered expert
-  reviewers that automatically adapt to your content.\n\n```ruby\ntext = \"The code
-  implements a binary search algorithm with O(log n) complexity\"\ncriteria = \"Evaluate
-  technical accuracy and clarity\"\n\nresult = ActiveGenie::Scoring.basic(text, criteria)\n#
+  Scoring\nText evaluation system that provides detailed scoring and feedback using
+  multiple expert reviewers. Get balanced scoring through AI-powered expert reviewers
+  that automatically adapt to your content.\n\n```ruby\ntext = \"The code implements
+  a binary search algorithm with O(log n) complexity\"\ncriteria = \"Evaluate technical
+  accuracy and clarity\"\n\nresult = ActiveGenie::Scoring.basic(\n  text,\n  criteria,\n
+  \ config: { provider: :anthropic, model: 'claude-3-5-haiku-20241022' } # optional\n)\n#
   => {\n#      algorithm_expert_score: 95,\n#      algorithm_expert_reasoning: \"Accurately
   describes binary search and its complexity\",\n#      technical_writer_score: 90,\n#
   \     technical_writer_reasoning: \"Clear and concise explanation of the algorithm\",\n#
-  \     final_score: 92.5\n#    }\n```\n\nFeatures:\n- Multi-reviewer evaluation with
-  automatic expert selection\n- Detailed feedback with scoring reasoning\n- Customizable
-  reviewer weights\n- Flexible evaluation criteria\n\nSee the [Scoring README](lib/active_genie/scoring/README.md)
+  \     final_score: 92.5\n#    }\n```\n\n*Recommended model*: `claude-3-5-haiku-20241022`\n\nFeatures:\n-
+  Multi-reviewer evaluation with automatic expert selection\n- Detailed feedback with
+  scoring reasoning\n- Customizable reviewer weights\n- Flexible evaluation criteria\n\nSee
+  the [Scoring README](lib/active_genie/scoring/README.md) for advanced usage, custom
+  reviewers, and detailed interface documentation.\n\n### Battle\nAI-powered battle
+  evaluation system that determines winners between two players based on specified
+  criteria.\n\n```ruby\nrequire 'active_genie'\n\nplayer_1 = \"Implementation uses
+  dependency injection for better testability\"\nplayer_2 = \"Code has high test coverage
+  but tightly coupled components\"\ncriteria = \"Evaluate code quality and maintainability\"\n\nresult
+  = ActiveGenie::Battle.call(\n  player_1,\n  player_2,\n  criteria,\n  config: {
+  provider: :google, model: 'gemini-2.0-flash-lite' } # optional\n)\n# => {\n#      winner_player:
+  \"Implementation uses dependency injection for better testability\",\n#      reasoning:
+  \"Player 1 implementation demonstrates better maintainability through dependency
+  injection, \n#                 which allows for easier testing and component replacement.
+  While Player 2 has good test coverage, \n#                 the tight coupling makes
+  the code harder to maintain and modify.\",\n#      what_could_be_changed_to_avoid_draw:
+  \"Focus on specific architectural patterns and design principles\"\n#    }\n```\n\n*Recommended
+  model*: `gemini-2.0-flash-lite`\n\nFeatures:\n- Multi-reviewer evaluation with automatic
+  expert selection\n- Detailed feedback with scoring reasoning\n- Customizable reviewer
+  weights\n- Flexible evaluation criteria\n\nSee the [Battle README](lib/active_genie/battle/README.md)
   for advanced usage, custom reviewers, and detailed interface documentation.\n\n###
-  Data Battle\nAI-powered battle evaluation system that determines winners between
-  two players based on specified criteria.\n\n```ruby\nrequire 'active_genie'\n\nplayer_a
-  = \"Implementation uses dependency injection for better testability\"\nplayer_b
-  = \"Code has high test coverage but tightly coupled components\"\ncriteria = \"Evaluate
-  code quality and maintainability\"\n\nresult = ActiveGenie::Battle.call(player_a,
-  player_b, criteria)\n# => {\n#      winner_player: \"Implementation uses dependency
-  injection for better testability\",\n#      reasoning: \"Player A's implementation
-  demonstrates better maintainability through dependency injection, \n#                 which
-  allows for easier testing and component replacement. While Player B has good test
-  coverage, \n#                 the tight coupling makes the code harder to maintain
-  and modify.\",\n#      what_could_be_changed_to_avoid_draw: \"Focus on specific
-  architectural patterns and design principles\"\n#    }\n```\n\nFeatures:\n- Multi-reviewer
-  evaluation with automatic expert selection\n- Detailed feedback with scoring reasoning\n-
-  Customizable reviewer weights\n- Flexible evaluation criteria\n\nSee the [Battle
-  README](lib/active_genie/battle/README.md) for advanced usage, custom reviewers,
-  and detailed interface documentation.\n\n### Data Ranking\nThe Ranking module provides
-  competitive ranking through multi-stage evaluation:\n\n\n```ruby\nrequire 'active_genie'\n\nplayers
-  = ['REST API', 'GraphQL API', 'SOAP API', 'gRPC API', 'Websocket API']\ncriteria
-  = \"Best one to be used into a high changing environment\"\n\nresult = ActiveGenie::Ranking.call(players,
-  criteria)\n# => {\n#      winner_player: \"gRPC API\",\n#      reasoning: \"gRPC
-  API is the best one to be used into a high changing environment\",\n#    }\n```\n\n-
+  Ranking\nThe Ranking module provides competitive ranking through multi-stage evaluation:\n\n```ruby\nrequire
+  'active_genie'\n\nplayers = ['REST API', 'GraphQL API', 'SOAP API', 'gRPC API',
+  'Websocket API']\ncriteria = \"Best one to be used into a high changing environment\"\n\nresult
+  = ActiveGenie::Ranking.call(\n  players,\n  criteria,\n  config: { provider: :google,
+  model: 'gemini-2.0-flash-lite' } # optional\n)\n# => {\n#      winner_player: \"gRPC
+  API\",\n#      reasoning: \"gRPC API is the best one to be used into a high changing
+  environment\",\n#    }\n```\n\n*Recommended model*: `gemini-2.0-flash-lite`\n\n-
   **Multi-phase ranking system** combining expert scoring and ELO algorithms\n- **Automatic
   elimination** of inconsistent performers using statistical analysis\n- **Dynamic
   ranking adjustments** based on simulated pairwise battles, from bottom to top\n\nSee
   the [Ranking README](lib/active_genie/ranking/README.md) for implementation details,
   configuration, and advanced ranking strategies.\n\n### Text Summarizer (Future)\n###
-  Language detector (Future)\n### Translator (Future)\n### Sentiment analyzer (Future)\n\n##
-  Configuration\n\n| Config | Description | Default |\n|--------|-------------|---------|\n|
-  `provider` | LLM provider (openai, anthropic, etc) | `nil` |\n| `model` | Model
-  to use | `nil` |\n| `api_key` | Provider API key | `nil` |\n| `timeout` | Request
-  timeout in seconds | `5` |\n| `max_retries` | Maximum retry attempts | `3` |\n\n>
-  **Note:** Each module can append its own set of configuration, see the individual
-  module documentation for details.\n\n## Contributing\n\n1. Fork the repository\n2.
-  Create your feature branch (`git checkout -b feature/amazing-feature`)\n3. Commit
-  your changes (`git commit -m 'Add amazing feature'`)\n4. Push to the branch (`git
-  push origin feature/amazing-feature`)\n5. Open a Pull Request\n\n## License\n\nThis
-  project is licensed under the Apache License 2.0 License - see the [LICENSE](LICENSE)
-  file for details.\n"
+  Categorizer (Future)\n### Language detector (Future)\n### Translator (Future)\n###
+  Sentiment analyzer (Future)\n\n## Benchmarking \U0001F9EA\n\nActiveGenie includes
+  a comprehensive benchmarking system to ensure consistent, high-quality outputs across
+  different LLM models and providers.\n\n```ruby\n# Run all benchmarks\nbundle exec
+  rake active_genie:benchmark\n\n# Run benchmarks for a specific module\nbundle exec
+  rake active_genie:benchmark[data_extractor]\n```\n\n### Latest Results\n\n| Model
+  | Overall Precision |\n|-------|-------------------|\n| claude-3-5-haiku-20241022
+  | 92.25% |\n| gemini-2.0-flash-lite | 84.25% |\n| gpt-4o-mini | 62.75% |\n| deepseek-chat
+  | 57.25% |\n\nSee the [Benchmark README](benchmark/README.md) for detailed results,
+  methodology, and how to contribute to our test suite.\n\n## Configuration\n\n| Config
+  | Description | Default |\n|--------|-------------|---------|\n| `provider` | LLM
+  provider (openai, anthropic, etc) | `nil` |\n| `model` | Model to use | `nil` |\n|
+  `api_key` | Provider API key | `nil` |\n| `timeout` | Request timeout in seconds
+  | `5` |\n| `max_retries` | Maximum retry attempts | `3` |\n\n> **Note:** Each module
+  can append its own set of configuration, see the individual module documentation
+  for details.\n\n## Contributing\n\n1. Fork the repository\n2. Create your feature
+  branch (`git checkout -b feature/amazing-feature`)\n3. Commit your changes (`git
+  commit -m 'Add amazing feature'`)\n4. Push to the branch (`git push origin feature/amazing-feature`)\n5.
+  Open a Pull Request\n\n## License\n\nThis project is licensed under the Apache License
+  2.0 License - see the [LICENSE](LICENSE) file for details.\n"
 email:
 - radames@roriz.dev
 executables: []
@@ -100,12 +111,21 @@ files:
 - lib/active_genie/battle.rb
 - lib/active_genie/battle/README.md
 - lib/active_genie/battle/basic.rb
+- lib/active_genie/clients/anthropic_client.rb
+- lib/active_genie/clients/google_client.rb
+- lib/active_genie/clients/helpers/retry.rb
 - lib/active_genie/clients/openai_client.rb
 - lib/active_genie/clients/unified_client.rb
+- lib/active_genie/concerns/loggable.rb
 - lib/active_genie/configuration.rb
 - lib/active_genie/configuration/log_config.rb
-- lib/active_genie/configuration/openai_config.rb
+- lib/active_genie/configuration/providers/anthropic_config.rb
+- lib/active_genie/configuration/providers/base_config.rb
+- lib/active_genie/configuration/providers/deepseek_config.rb
+- lib/active_genie/configuration/providers/google_config.rb
+- lib/active_genie/configuration/providers/openai_config.rb
 - lib/active_genie/configuration/providers_config.rb
+- lib/active_genie/configuration/runtime_config.rb
 - lib/active_genie/data_extractor.rb
 - lib/active_genie/data_extractor/README.md
 - lib/active_genie/data_extractor/basic.rb
@@ -123,6 +143,7 @@ files:
 - lib/active_genie/scoring/README.md
 - lib/active_genie/scoring/basic.rb
 - lib/active_genie/scoring/recommended_reviewers.rb
+- lib/tasks/benchmark.rake
 - lib/tasks/install.rake
 - lib/tasks/templates/active_genie.rb
 homepage: https://github.com/Roriz/active_genie
@@ -142,7 +163,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.0.0
+      version: 3.0.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="

data/lib/active_genie/configuration/openai_config.rb

@@ -1,56 +0,0 @@
-require_relative '../clients/openai_client'
-
-module ActiveGenie::Configuration
-  class OpenaiConfig
-    attr_writer :api_key, :organization, :api_url, :client,
-      :lower_tier_model, :middle_tier_model, :upper_tier_model
-
-    def api_key
-      @api_key || ENV['OPENAI_API_KEY']
-    end
-
-    def organization
-      @organization || ENV['OPENAI_ORGANIZATION']
-    end
-
-    def api_url
-      @api_url || 'https://api.openai.com/v1'
-    end
-
-    def client
-      @client ||= ::ActiveGenie::Clients::OpenaiClient.new(self)
-    end
-
-    def lower_tier_model
-      @lower_tier_model || 'gpt-4o-mini'
-    end
-
-    def middle_tier_model
-      @middle_tier_model || 'gpt-4o'
-    end
-
-    def upper_tier_model
-      @upper_tier_model || 'o1-preview'
-    end
-
-    def tier_to_model(tier)
-      {
-        lower_tier: lower_tier_model,
-        middle_tier: middle_tier_model,
-        upper_tier: upper_tier_model
-      }[tier&.to_sym] || lower_tier_model
-    end
-
-    def to_h(config = {})
-      {
-        api_key:,
-        organization:,
-        api_url:,
-        lower_tier_model:,
-        middle_tier_model:,
-        upper_tier_model:,
-        **config
-      }
-    end
-  end
-end