dspy 0.28.2 → 0.29.1

data/lib/gepa/utils/pareto.rb

@@ -0,0 +1,119 @@
+# frozen_string_literal: true
+
+require 'json'
+require 'set'
+require 'sorbet-runtime'
+
+module GEPA
+  module Utils
+    module Pareto
+      extend T::Sig
+
+      sig { params(value: T.untyped).returns(T.untyped) }
+      def self.json_default(value)
+        value.is_a?(Hash) ? value.transform_keys(&:to_s) : JSON.parse(value.to_json)
+      rescue StandardError
+        { value: value.to_s }
+      end
+
+      sig { params(values: T::Array[Float]).returns(Integer) }
+      def self.idxmax(values)
+        raise ArgumentError, 'values must not be empty' if values.empty?
+
+        values.each_with_index.max_by { |score, _i| score }&.last || 0
+      end
+
+      sig do
+        params(
+          program_at_pareto_front_valset: T::Array[T.untyped],
+          scores: T.nilable(T::Hash[Integer, Float])
+        ).returns(T::Array[T.untyped])
+      end
+      def self.remove_dominated_programs(program_at_pareto_front_valset, scores: nil)
+        normalized_fronts = program_at_pareto_front_valset.map { |front| front.to_a }
+
+        frequency = Hash.new(0)
+        normalized_fronts.each do |front|
+          front.each { |program_idx| frequency[program_idx] += 1 }
+        end
+
+        all_programs = frequency.keys
+        scores ||= all_programs.to_h { |idx| [idx, 1.0] }
+
+        sorted_programs = all_programs.sort_by { |idx| scores.fetch(idx, 0.0) }
+
+        dominated = Set.new
+        loop do
+          found = false
+          sorted_programs.each do |candidate|
+            next if dominated.include?(candidate)
+            next unless dominated?(candidate, sorted_programs.to_set, dominated, normalized_fronts)
+
+            dominated.add(candidate)
+            found = true
+            break
+          end
+          break unless found
+        end
+
+        dominators = sorted_programs.reject { |idx| dominated.include?(idx) }
+        dominators_set = dominators.to_set
+
+        normalized_fronts.map do |front|
+          front.select { |idx| dominators_set.include?(idx) }
+        end
+      end
+
+      sig do
+        params(
+          pareto_front_programs: T::Array[T.untyped],
+          train_val_weighted_scores: T::Hash[Integer, Float]
+        ).returns(T::Array[Integer])
+      end
+      def self.find_dominator_programs(pareto_front_programs, train_val_weighted_scores)
+        cleaned_frontiers = remove_dominated_programs(pareto_front_programs, scores: train_val_weighted_scores)
+        cleaned_frontiers.flat_map(&:to_a).uniq
+      end
+
+      sig do
+        params(
+          pareto_front_programs: T::Array[T.untyped],
+          weighted_scores: T::Hash[Integer, Float],
+          rng: Random
+        ).returns(Integer)
+      end
+      def self.select_program_candidate_from_pareto_front(pareto_front_programs, weighted_scores, rng)
+        cleaned_frontiers = remove_dominated_programs(pareto_front_programs, scores: weighted_scores)
+        frequency = Hash.new(0)
+        cleaned_frontiers.each do |front|
+          front.each { |idx| frequency[idx] += 1 }
+        end
+        raise ArgumentError, 'pareto front is empty' if frequency.empty?
+
+        sampling_list = frequency.flat_map { |idx, freq| [idx] * freq }
+        sampling_list[rng.rand(sampling_list.length)]
+      end
+
+      class << self
+        extend T::Sig
+        private
+
+        sig do
+          params(
+            candidate: Integer,
+            program_set: Set,
+            dominated: Set,
+            pareto_fronts: T::Array[T::Array[Integer]]
+          ).returns(T::Boolean)
+        end
+        def dominated?(candidate, program_set, dominated, pareto_fronts)
+          candidate_fronts = pareto_fronts.select { |front| front.include?(candidate) }
+          candidate_fronts.all? do |front|
+            remaining = front.reject { |idx| idx == candidate || dominated.include?(idx) }
+            remaining.any? { |other| program_set.include?(other) }
+          end
+        end
+      end
+    end
+  end
+end

data/lib/gepa.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+require_relative 'gepa/telemetry'
+require_relative 'gepa/logging'
+require_relative 'gepa/utils/pareto'
+require_relative 'gepa/strategies/batch_sampler'
+require_relative 'gepa/strategies/candidate_selector'
+require_relative 'gepa/strategies/component_selector'
+require_relative 'gepa/strategies/instruction_proposal'
+require_relative 'gepa/core/evaluation_batch'
+require_relative 'gepa/core/result'
+require_relative 'gepa/core/state'
+require_relative 'gepa/core/engine'
+require_relative 'gepa/proposer/base'
+require_relative 'gepa/proposer/reflective_mutation/base'
+require_relative 'gepa/proposer/reflective_mutation/reflective_mutation'
+require_relative 'gepa/proposer/merge_proposer'
+require_relative 'gepa/api'
+
+module GEPA
+end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: dspy
 version: !ruby/object:Gem::Version
-  version: 0.28.2
+  version: 0.29.1
 platform: ruby
 authors:
 - Vicente Reig Rincón de Arellano
 bindir: bin
 cert_chain: []
-date: 2025-10-13 00:00:00.000000000 Z
+date: 2025-10-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: dry-configurable
@@ -51,6 +51,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '2.29'
+- !ruby/object:Gem::Dependency
+  name: concurrent-ruby
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.3'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.3'
 - !ruby/object:Gem::Dependency
   name: openai
   requirement: !ruby/object:Gem::Requirement
@@ -207,6 +221,8 @@ files:
 - lib/dspy/chain_of_thought.rb
 - lib/dspy/code_act.rb
 - lib/dspy/context.rb
+- lib/dspy/datasets.rb
+- lib/dspy/datasets/ade.rb
 - lib/dspy/error_formatter.rb
 - lib/dspy/errors.rb
 - lib/dspy/evaluate.rb
@@ -257,6 +273,7 @@ files:
 - lib/dspy/propose/dataset_summary_generator.rb
 - lib/dspy/propose/grounded_proposer.rb
 - lib/dspy/re_act.rb
+- lib/dspy/reflection_lm.rb
 - lib/dspy/registry/registry_manager.rb
 - lib/dspy/registry/signature_registry.rb
 - lib/dspy/schema_adapters.rb
@@ -268,7 +285,6 @@ files:
 - lib/dspy/teleprompt/data_handler.rb
 - lib/dspy/teleprompt/gepa.rb
 - lib/dspy/teleprompt/mipro_v2.rb
-- lib/dspy/teleprompt/simple_optimizer.rb
 - lib/dspy/teleprompt/teleprompter.rb
 - lib/dspy/teleprompt/utils.rb
 - lib/dspy/tools.rb
@@ -281,6 +297,25 @@ files:
 - lib/dspy/type_system/sorbet_json_schema.rb
 - lib/dspy/utils/serialization.rb
 - lib/dspy/version.rb
+- lib/gepa.rb
+- lib/gepa/api.rb
+- lib/gepa/core/engine.rb
+- lib/gepa/core/evaluation_batch.rb
+- lib/gepa/core/result.rb
+- lib/gepa/core/state.rb
+- lib/gepa/logging.rb
+- lib/gepa/logging/experiment_tracker.rb
+- lib/gepa/logging/logger.rb
+- lib/gepa/proposer/base.rb
+- lib/gepa/proposer/merge_proposer.rb
+- lib/gepa/proposer/reflective_mutation/base.rb
+- lib/gepa/proposer/reflective_mutation/reflective_mutation.rb
+- lib/gepa/strategies/batch_sampler.rb
+- lib/gepa/strategies/candidate_selector.rb
+- lib/gepa/strategies/component_selector.rb
+- lib/gepa/strategies/instruction_proposal.rb
+- lib/gepa/telemetry.rb
+- lib/gepa/utils/pareto.rb
 homepage: https://github.com/vicentereig/dspy.rb
 licenses:
 - MIT