disco 0.4.0 → 0.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0fb469aae804d8f6ed2fe398b3584e67d1ebf29cd0d0d394eda952d32c0756f9
-  data.tar.gz: 295bde072449c13e3da22975efa0bdbe9ea5a1fdbbb787f0b0d63411056e6799
+  metadata.gz: eb58ae5c2579ef03fc0089f9f9e0421cd38bd8996472bf04ef9383b601759e5a
+  data.tar.gz: 9bb8c5aa186ab1e6364d0ed446dc61e69a0f0ee730cee4c8367586dac69b666a
 SHA512:
-  metadata.gz: 287c44da295f55d2f95788fb7c0488738c1f70a409bb372461ad37d1786ade8fd0650abb8de62f023805a03bd667f74b509069bbaa9ca4f58a56aa1628d8b7c4
-  data.tar.gz: 81bceaa14ffc78918157b25c4cb4a21f0d08d625042867866032fe006a8cbfcd041914dfa19c720a0bc7211d00fe40d14f95fc2e1d769b7a9af59711641dcc78
+  metadata.gz: ecee89ddb2db25a9ba697ec637f16a2408079b8b3d7d287df0cfdb30d9b7c0a996d6cb38075766d7e3fecf529be6ad90187a1ec95d2d141076ecfe3e2236209a
+  data.tar.gz: 31d7259ac86779530b468392b611f0920b24682f8017e3b50f9a3ca41d30ad08454219881e26eae67d1add7a47d645f7a8b77d253ada4a3d0b5db0f5dbf5e33d

data/CHANGELOG.md

@@ -1,3 +1,7 @@
+## 0.4.1 (2024-05-23)
+
+- Reduced memory for `item_recs` and `similar_users`
+
 ## 0.4.0 (2023-01-30)
 
 - Fixed issue with `has_recommended` and inheritance with Rails < 6.1

data/README.md

@@ -6,7 +6,7 @@
 - Works with explicit and implicit feedback
 - Uses high-performance matrix factorization
 
-[![Build Status](https://github.com/ankane/disco/workflows/build/badge.svg?branch=master)](https://github.com/ankane/disco/actions)
+[![Build Status](https://github.com/ankane/disco/actions/workflows/build.yml/badge.svg)](https://github.com/ankane/disco/actions)
 
 ## Installation
 
@@ -229,8 +229,8 @@ recommender.user_recs(new_user_id) # returns empty array
 
 There are a number of ways to deal with this, but here are some common ones:
 
-- For user-based recommendations, show new users the most popular items.
-- For item-based recommendations, make content-based recommendations with a gem like [tf-idf-similarity](https://github.com/jpmckinney/tf-idf-similarity).
+- For user-based recommendations, show new users the most popular items
+- For item-based recommendations, make content-based recommendations with a gem like [tf-idf-similarity](https://github.com/jpmckinney/tf-idf-similarity)
 
 Get top items with:
 
@@ -329,28 +329,6 @@ Thanks to:
 - [Implicit](https://github.com/benfred/implicit/) for serving as an initial reference for user and item similarity
 - [@dasch](https://github.com/dasch) for the gem name
 
-## Upgrading
-
-### 0.2.7
-
-There’s now a warning when passing `:value` with implicit feedback, as this has no effect on recommendations and can be removed. Earlier versions of the library incorrectly stated this was used.
-
-```ruby
-recommender.fit([
-  {user_id: 1, item_id: 1, value: 1},
-  {user_id: 2, item_id: 1, value: 3}
-])
-```
-
-to:
-
-```ruby
-recommender.fit([
-  {user_id: 1, item_id: 1},
-  {user_id: 2, item_id: 1}
-])
-```
-
 ## History
 
 View the [changelog](https://github.com/ankane/disco/blob/master/CHANGELOG.md)

data/lib/disco/data.rb

@@ -9,7 +9,7 @@ module Disco
         file_hash: "06416e597f82b7342361e41163890c81036900f418ad91315590814211dca490")
 
       # convert u.item to utf-8
-      movies_str = File.read(item_path).encode("UTF-8", "binary", invalid: :replace, undef: :replace, replace: "")
+      movies_str = File.read(item_path).encode("UTF-8", "ISO-8859-1")
 
       movies = {}
       CSV.parse(movies_str, col_sep: "|") do |row|

data/lib/disco/recommender.rb

@@ -99,8 +99,8 @@ module Disco
       @user_factors = model.p_factors(format: :numo)
       @item_factors = model.q_factors(format: :numo)
 
-      @normalized_user_factors = nil
-      @normalized_item_factors = nil
+      @user_norms = nil
+      @item_norms = nil
 
       @user_recs_index = nil
       @similar_users_index = nil
@@ -172,13 +172,13 @@ module Disco
 
     def similar_items(item_id, count: 5)
       check_fit
-      similar(item_id, :item_id, @item_map, normalized_item_factors, count, @similar_items_index)
+      similar(item_id, :item_id, @item_map, @item_factors, item_norms, count, @similar_items_index)
     end
     alias_method :item_recs, :similar_items
 
     def similar_users(user_id, count: 5)
       check_fit
-      similar(user_id, :user_id, @user_map, normalized_user_factors, count, @similar_users_index)
+      similar(user_id, :user_id, @user_map, @user_factors, user_norms, count, @similar_users_index)
     end
 
     def top_items(count: 5)
@@ -247,13 +247,13 @@ module Disco
 
     def optimize_similar_items(library: nil)
       check_fit
-      @similar_items_index = create_index(normalized_item_factors, library: library)
+      @similar_items_index = create_index(@item_factors / item_norms.expand_dims(1), library: library)
     end
     alias_method :optimize_item_recs, :optimize_similar_items
 
     def optimize_similar_users(library: nil)
       check_fit
-      @similar_users_index = create_index(normalized_user_factors, library: library)
+      @similar_users_index = create_index(@user_factors / user_norms.expand_dims(1), library: library)
     end
 
     def inspect
@@ -341,36 +341,37 @@ module Disco
       end
     end
 
-    def normalized_user_factors
-      @normalized_user_factors ||= normalize(@user_factors)
+    def user_norms
+      @user_norms ||= norms(@user_factors)
     end
 
-    def normalized_item_factors
-      @normalized_item_factors ||= normalize(@item_factors)
+    def item_norms
+      @item_norms ||= norms(@item_factors)
     end
 
-    def normalize(factors)
+    def norms(factors)
       norms = Numo::SFloat::Math.sqrt((factors * factors).sum(axis: 1))
       norms[norms.eq(0)] = 1e-10 # no zeros
-      factors / norms.expand_dims(1)
+      norms
     end
 
-    def similar(id, key, map, norm_factors, count, index)
+    def similar(id, key, map, factors, norms, count, index)
       i = map[id]
 
-      if i && norm_factors.shape[0] > 1
+      if i && factors.shape[0] > 1
         if index && count
+          norm_factors = factors[i, true] / norms[i]
           if defined?(Faiss) && index.is_a?(Faiss::Index)
-            predictions, ids = index.search(norm_factors[i, true].expand_dims(0), count + 1).map { |v| v.to_a[0] }
+            predictions, ids = index.search(norm_factors.expand_dims(0), count + 1).map { |v| v.to_a[0] }
           else
-            result = index.search(norm_factors[i, true], size: count + 1)
+            result = index.search(norm_factors, size: count + 1)
             # ids from batch_insert start at 1 instead of 0
             ids = result.map { |v| v[:id] - 1 }
             # convert cosine distance to cosine similarity
             predictions = result.map { |v| 1 - v[:distance] }
           end
         else
-          predictions = norm_factors.inner(norm_factors[i, true])
+          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]
@@ -386,6 +387,7 @@ module Disco
           next if id == i
 
           result << {key => keys[id], score: predictions[j]}
+          break if result.size == count
         end
         result
       else

data/lib/disco/version.rb

@@ -1,3 +1,3 @@
 module Disco
-  VERSION = "0.4.0"
+  VERSION = "0.4.1"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: disco
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.4.1
 platform: ruby
 authors:
 - Andrew Kane
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-01-30 00:00:00.000000000 Z
+date: 2024-05-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: libmf
@@ -76,7 +76,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.1
+rubygems_version: 3.5.9
 signing_key:
 specification_version: 4
 summary: Recommendations for Ruby and Rails using collaborative filtering