disco 0.5.0 → 0.5.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ffa944ca55d7da5a4dc4cced181d013a9fdc0f1fbb76af6003ba3049bb299682
-  data.tar.gz: 64f389a3a93173b3d24cad85e5072b4d3f8965a729387a5845ed9c86098a9496
+  metadata.gz: 8ca6da20099f6ec36c5f07547c61bb368066f2f480fba03c775287f48407a1f8
+  data.tar.gz: 3fbd73bb42398534d31a7b574d5d8dd6e40a1fe867e7cf4fe0ede4bdad5cd0c9
 SHA512:
-  metadata.gz: f167237c7553d5a506467844d6acf860e81efca11da4e8facb8988cc994fbe70a52b38ff87b0d1157bcf9312b172c8794b49e6e8607748f77b129bcac26742d2
-  data.tar.gz: f5feab5c5d88dac8cc5a01cbab437110d0fe41c4fa98283464174d253996d8c8efa9dc92724f30bbae334c7fe2c7cb932b2b6335bd37f44aa645ef33fa7ec7ea
+  metadata.gz: 570275aaf7887c98318040efde2440f5b6e8e04be25faa7a78b73453e05dec845963067f0808eff09e28621dbd7b3277e98da83521e067b6170a585f717462f5
+  data.tar.gz: 57b86344de0656aa8e4bb0f130d0f3bcd928d05365b7f39479e4c6a69cd49d3a2f2a10082abedaa351d5febb212681236548520a01c9b321132e6f798e74a690

data/CHANGELOG.md

@@ -1,3 +1,12 @@
+## 0.5.2 (2025-09-12)
+
+- Fixed recommendations when numo-narray compiled with GCC 13+
+- Fixed error with Rover 0.5+
+
+## 0.5.1 (2024-12-29)
+
+- Removed dependency on `base64` gem for serialization
+
 ## 0.5.0 (2024-10-22)
 
 - Changed dataset directory to match XDG Base Directory Specification

data/lib/disco/recommender.rb

@@ -138,17 +138,13 @@ module Disco
           return [] if ids.size == 0
 
           predictions = @item_factors[ids, true].inner(@user_factors[u, true])
-          indexes = predictions.sort_index.reverse
-          indexes = indexes[0...[count + rated.size, indexes.size].min] if count
-          predictions = predictions[indexes]
+          predictions, indexes = top_k(predictions, count ? count + rated.size : nil)
           ids = ids[indexes]
         elsif @user_recs_index && count
           predictions, ids = @user_recs_index.search(@user_factors[u, true].expand_dims(0), count + rated.size).map { |v| v[0, true] }
         else
           predictions = @item_factors.inner(@user_factors[u, true])
-          indexes = predictions.sort_index.reverse # reverse just creates view
-          indexes = indexes[0...[count + rated.size, indexes.size].min] if count
-          predictions = predictions[indexes]
+          predictions, indexes = top_k(predictions, count ? count + rated.size : nil)
           ids = indexes
         end
 
@@ -196,7 +192,7 @@ module Disco
         # wilson score with continuity correction
         # https://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval#Wilson_score_interval_with_continuity_correction
         z = 1.96 # 95% confidence
-        range = @max_rating - @min_rating
+        range = @max_rating - min_rating
         n = Numo::DFloat.cast(@item_count)
         phat = (Numo::DFloat.cast(@item_sum) - (min_rating * n)) / range / n
         phat = (phat - (1 / (2 * n))).clip(0, nil) # continuity correction
@@ -204,9 +200,7 @@ module Disco
         scores = scores * range + min_rating
       end
 
-      indexes = scores.sort_index.reverse
-      indexes = indexes[0...[count, indexes.size].min] if count
-      scores = scores[indexes]
+      scores, indexes = top_k(scores, count)
 
       keys = @item_map.keys
       indexes.size.times.map do |i|
@@ -261,7 +255,6 @@ module Disco
     end
 
     def to_json
-      require "base64"
       require "json"
 
       obj = {
@@ -270,8 +263,8 @@ module Disco
         item_ids: @item_map.keys,
         rated: @user_map.map { |_, u| (@rated[u] || {}).keys },
         global_mean: @global_mean,
-        user_factors: Base64.strict_encode64(@user_factors.to_binary),
-        item_factors: Base64.strict_encode64(@item_factors.to_binary),
+        user_factors: [@user_factors.to_binary].pack("m0"),
+        item_factors: [@item_factors.to_binary].pack("m0"),
         factors: @factors,
         epochs: @epochs,
         verbose: @verbose
@@ -372,9 +365,7 @@ module Disco
           end
         else
           predictions = factors.inner(factors[i, true]) / (norms * norms[i])
-          indexes = predictions.sort_index.reverse
-          indexes = indexes[0...[count + 1, indexes.size].min] if count
-          predictions = predictions[indexes]
+          predictions, indexes = top_k(predictions, count ? count + 1 : nil)
           ids = indexes
         end
 
@@ -395,6 +386,27 @@ module Disco
       end
     end
 
+    def top_k(values, count)
+      if self.class.sort_index?
+        indexes = values.sort_index.reverse
+        indexes = indexes[0...[count, indexes.size].min] if count
+      else
+        indexes = values.to_a.each_with_index.sort_by { |v, _| -v }
+        indexes = indexes.first(count) if count
+        indexes = indexes.map(&:last)
+      end
+      [values[indexes], indexes]
+    end
+
+    # https://github.com/ruby-numo/numo-narray/issues/243
+    def self.sort_index?
+      unless defined?(@sort_index)
+        arr = Numo::SFloat.new(100).rand
+        @sort_index = arr[arr.sort_index].to_a == arr.to_a.sort
+      end
+      @sort_index
+    end
+
     def check_ratings(ratings)
       unless ratings.all? { |r| !r[:rating].nil? }
         raise ArgumentError, "Missing rating"
@@ -415,11 +427,7 @@ module Disco
     def to_dataset(dataset)
       if defined?(Rover::DataFrame) && dataset.is_a?(Rover::DataFrame)
         # convert keys to symbols
-        dataset = dataset.dup
-        dataset.keys.each do |k, v|
-          dataset[k.to_sym] ||= dataset.delete(k)
-        end
-        dataset.to_a
+        dataset.each_row.map { |v| v.transform_keys(&:to_sym) }
       elsif defined?(Daru::DataFrame) && dataset.is_a?(Daru::DataFrame)
         # convert keys to symbols
         dataset = dataset.dup
@@ -432,16 +440,14 @@ module Disco
     end
 
     def json_load(obj)
-      require "base64"
-
       @implicit = obj["implicit"]
       @user_map = obj["user_ids"].map.with_index.to_h
       @item_map = obj["item_ids"].map.with_index.to_h
       @rated = obj["rated"].map.with_index.to_h { |r, i| [i, r.to_h { |v| [v, true] }] }
       @global_mean = obj["global_mean"].to_f
       @factors = obj["factors"].to_i
-      @user_factors = Numo::SFloat.from_binary(Base64.strict_decode64(obj["user_factors"]), [@user_map.size, @factors])
-      @item_factors = Numo::SFloat.from_binary(Base64.strict_decode64(obj["item_factors"]), [@item_map.size, @factors])
+      @user_factors = Numo::SFloat.from_binary(obj["user_factors"].unpack1("m0"), [@user_map.size, @factors])
+      @item_factors = Numo::SFloat.from_binary(obj["item_factors"].unpack1("m0"), [@item_map.size, @factors])
       @epochs = obj["epochs"].to_i
       @verbose = obj["verbose"]
 

data/lib/disco/version.rb

@@ -1,3 +1,3 @@
 module Disco
-  VERSION = "0.5.0"
+  VERSION = "0.5.2"
 end

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: disco
 version: !ruby/object:Gem::Version
-  version: 0.5.0
+  version: 0.5.2
 platform: ruby
 authors:
 - Andrew Kane
-autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-10-23 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: libmf
@@ -38,7 +37,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 0.9.2
-description:
 email: andrew@ankane.org
 executables: []
 extensions: []
@@ -61,7 +59,6 @@ homepage: https://github.com/ankane/disco
 licenses:
 - MIT
 metadata: {}
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -76,8 +73,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.5.16
-signing_key:
+rubygems_version: 3.6.9
 specification_version: 4
 summary: Recommendations for Ruby and Rails using collaborative filtering
 test_files: []