easy_ml 0.2.0.pre.rc82 → 0.2.0.pre.rc84

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ce245d6900c4c5c001c0de9982894ccf6b41faef31e8c958dc540ef05fe426e4
-  data.tar.gz: a120f14076a9ff83ca6afb8b0bd651b9ea9ed0185d42f23b82f5de6b2a4de831
+  metadata.gz: ff472f1d00b1629047b7b414ae6d8104bc34529602808d919e5117ba4fd99077
+  data.tar.gz: 676ba30c99251193848aa2f02957e18d1abc9fdfef112ab0699720e8edc96d1a
 SHA512:
-  metadata.gz: db2b7292bf07b5122a7949a111c56a25f8848496eaadfc94f101a271054fd4c21d401b4c941f69d7579c8cdc9887e68c4105f11720dba13aa7fd0fcdafb79b81
-  data.tar.gz: f5dc8ffee52fb67b6cda8e4462e540900d528dd9c15ff49faa8ca45f21f9e4968129b996a5018a91801aae431a65b5274d64e69045e5feeeec6f12f09900cda3
+  metadata.gz: a69606a0cf1a10939c5fa29ac8425e2cfbec08dc99e4c0dda510b7cc227171654a89f11ac9cba987636e5cf2fa7ada7d98e179dc04f880514fbf6013f2f31bda
+  data.tar.gz: 5bf4737b1d4fa4dcefa999be84e76152c3a665826a5e55d72a9a120fbf5dd79cf53a44a4f38dc78fa6abc40f38e1c13779c0ab7b98f219fe29c29005fd1e5134

data/app/jobs/easy_ml/training_job.rb

@@ -10,13 +10,13 @@ module EasyML
 
       @last_activity = Time.current
       setup_signal_traps
-      # @monitor_thread = start_monitor_thread
+      @monitor_thread = start_monitor_thread
 
       @model.actually_train do |iteration_info|
         @last_activity = Time.current
       end
     ensure
-      # @monitor_thread&.exit
+      @monitor_thread&.exit
       @model.unlock!
     end
 

data/app/models/easy_ml/feature.rb

@@ -34,6 +34,7 @@ module EasyML
       def compute_sha(feature_class)
         require "digest"
         path = feature_class.constantize.instance_method(:transform).source_location.first
+        return nil unless File.exist?(path)
         current_mtime = File.mtime(path)
         cache_key = "feature_sha/#{path}"
 

data/lib/easy_ml/data/dataset_manager/writer/base.rb

@@ -95,7 +95,7 @@ module EasyML
             keylist = unique_id_key(subdir: "keylist")
 
             acquire_lock(keylist) do |suo|
-              suo.client.sadd?(keylist, key)
+              suo.client.sadd(keylist, key)
             end
           end
 

data/lib/easy_ml/version.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 module EasyML
-  VERSION = "0.2.0-rc82"
+  VERSION = "0.2.0-rc84"
 
   module Version
   end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: easy_ml
 version: !ruby/object:Gem::Version
-  version: 0.2.0.pre.rc82
+  version: 0.2.0.pre.rc84
 platform: ruby
 authors:
 - Brett Shollenberger
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2025-02-20 00:00:00.000000000 Z
+date: 2025-02-24 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activerecord
@@ -719,9 +719,6 @@ files:
 - lib/easy_ml/data/dataset_manager/writer/partitioned.rb
 - lib/easy_ml/data/dataset_manager/writer/partitioned/partition_reasons.rb
 - lib/easy_ml/data/date_converter.rb
-- lib/easy_ml/data/embeddings.rb
-- lib/easy_ml/data/embeddings/adapters.rb
-- lib/easy_ml/data/embeddings/compression.rb
 - lib/easy_ml/data/partition.rb
 - lib/easy_ml/data/partition/boundaries.rb
 - lib/easy_ml/data/polars_column.rb

data/lib/easy_ml/data/embeddings/adapters.rb

@@ -1,56 +0,0 @@
-module EasyML
-  module Data
-    class Embeddings
-      class Adapters
-        attr_accessor :model, :config
-
-        ADAPTERS = {
-          anthropic: Langchain::LLM::Anthropic,
-          gemini: Langchain::LLM::GoogleGemini,
-          openai: Langchain::LLM::OpenAI,
-          ollama: Langchain::LLM::Ollama,
-        }
-
-        DEFAULTS = {
-          api_key: {
-            anthropic: ENV["ANTHROPIC_API_KEY"],
-            gemini: ENV["GEMINI_API_KEY"],
-            openai: ENV["OPENAI_API_KEY"],
-            ollama: ENV["OLLAMA_API_KEY"],
-          },
-        }
-
-        def initialize(model, config = {})
-          @model = model.to_sym
-          @config = config.symbolize_keys
-          apply_defaults
-        end
-
-        def embed(df, col)
-          pick
-          texts = df[col].to_a
-          df = df.with_column(
-            embeddings: adapter.embed(text: texts),
-          )
-        end
-
-        private
-
-        def pick
-          @adapter ||= ADAPTERS[@model].new(config)
-          self
-        end
-
-        def apply_defaults
-          @config = @config.deep_symbolize_keys
-
-          DEFAULTS.each do |k, v|
-            unless @config.key?(k)
-              @config[k] = v[@model]
-            end
-          end
-        end
-      end
-    end
-  end
-end

data/lib/easy_ml/data/embeddings/compression.rb

@@ -1 +0,0 @@
-

data/lib/easy_ml/data/embeddings.rb

@@ -1,43 +0,0 @@
-module EasyML
-  module Data
-    class Embeddings
-      COMPRESSION_DEFAULT = {
-        present: :balanced,
-      }
-
-      attr_reader :df, :column, :model, :adapter, :compression,
-                  :embeddings, :compressed_embeddings
-
-      def initialize(options = {})
-        @df = options[:df]
-        @column = options[:column]
-        @model = options[:model]
-        @config = options[:config] || {}
-        @compression = options[:compression] || COMPRESSION_DEFAULT
-      end
-
-      def create
-        embed
-        compress
-      end
-
-      def embed
-        @embeddings ||= adapter.embed(df, column)
-      end
-
-      def compress
-        @compressed_embeddings ||= compression_adapter.compress(embeddings)
-      end
-
-      private
-
-      def adapter
-        @adapter ||= EasyML::Data::Embeddings::Adapters.new(model, config)
-      end
-
-      def compression_adapter
-        @compression_adapter ||= EasyML::Data::Embeddings::Compression.new(compression)
-      end
-    end
-  end
-end