libmf 0.2.6 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 33b8ba1c7cf9de8e19fa9775ecdb445bc0f2f4323401cccbf016ff6a81a04a00
-  data.tar.gz: 9ff88bdafac8bbfe0c99ccc0c0b7dfb7a9092246de1105c7f99ccade29b9c88b
+  metadata.gz: b264767ceb43c4a0407896e2bcf2373a47296e1e0580c65ec2ef08471589f661
+  data.tar.gz: addfb47cc44989620f847eac0392aef7431e596af08095ddd56ed227c4b99274
 SHA512:
-  metadata.gz: 021eca76e96ceca2cd484b7f7796a649529f3b20c80d494143cfaf43e617d23ac55ae50bdcd345808e277b31d304ea6e6a53c937b3e9a200f47ae226f27e4f2b
-  data.tar.gz: 916b78a5e62d19e4d6a9e1433350bb26982d43393029e4e883ed803f89a03f6db766ab10718120a73db1f0ad7ac80607d475a10a7c203d117ca1706e80618345
+  metadata.gz: b22bcab875b5c2e8b7432c137f90b3ea37a8c16c3e64aa0fb4c15f824f8345a3105ca981a4ea1ea091396dcd0ce10320eab52f56e6ede75601ea2192e5bc61ac
+  data.tar.gz: 2e946522f35f77a46ab3880d74da38a0efdcaf432eabc1945d35af344d5cafb2c7712a3e0660352f8cf84fbd44af5c9c4e1cc6110ef5b908cd66c3393e124a99

data/CHANGELOG.md

@@ -1,3 +1,14 @@
+## 0.4.0 (2024-10-22)
+
+- Dropped support for Ruby < 3.1
+
+## 0.3.0 (2022-08-07)
+
+- Added metrics
+- Prefer `save` over `save_model`
+- Prefer `load` over `load_model`
+- Dropped support for Ruby < 2.7
+
 ## 0.2.6 (2021-12-02)
 
 - Improved ARM detection

data/LICENSE.txt

@@ -1,30 +1,33 @@
 BSD 3-Clause License
 
-Copyright (c) 2014-2015, The LIBMF Project
-Copyright (c) 2019-2021, Andrew Kane
+Copyright (c) 2014-2015 The LIBMF Project.
+Copyright (c) 2019-2024 Andrew Kane.
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without
-modification, are permitted provided that the following conditions are met:
+modification, are permitted provided that the following conditions
+are met:
 
-1. Redistributions of source code must retain the above copyright notice, this
-   list of conditions and the following disclaimer.
+1. Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
 
-2. Redistributions in binary form must reproduce the above copyright notice,
-   this list of conditions and the following disclaimer in the documentation
-   and/or other materials provided with the distribution.
+2. Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
 
-3. Neither the name of the copyright holder nor the names of its
-   contributors may be used to endorse or promote products derived from
-   this software without specific prior written permission.
+3. Neither name of copyright holders nor the names of its contributors
+may be used to endorse or promote products derived from this software
+without specific prior written permission.
 
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
-AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
-DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
-FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
-DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
-SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
-OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

data/README.md

@@ -4,14 +4,14 @@
 
 Check out [Disco](https://github.com/ankane/disco) for higher-level collaborative filtering
 
-[![Build Status](https://github.com/ankane/libmf-ruby/workflows/build/badge.svg?branch=master)](https://github.com/ankane/libmf-ruby/actions)
+[![Build Status](https://github.com/ankane/libmf-ruby/actions/workflows/build.yml/badge.svg)](https://github.com/ankane/libmf-ruby/actions)
 
 ## Installation
 
 Add this line to your application’s Gemfile:
 
 ```ruby
-gem 'libmf'
+gem "libmf"
 ```
 
 ## Getting Started
@@ -54,13 +54,13 @@ model.bias
 Save the model to a file
 
 ```ruby
-model.save_model("model.txt")
+model.save("model.txt")
 ```
 
 Load the model from a file
 
 ```ruby
-model.load_model("model.txt")
+model = Libmf::Model.load("model.txt")
 ```
 
 Pass a validation set
@@ -99,7 +99,7 @@ Libmf::Model.new(
   lambda_q1: 0,           # coefficient of L1-norm regularization on Q
   lambda_q2: 0.1,         # coefficient of L2-norm regularization on Q
   learning_rate: 0.1,     # learning rate
-  alpha: 0.1,             # importance of negative entries
+  alpha: 1,               # importance of negative entries
   c: 0.0001,              # desired value of negative entries
   nmf: false,             # perform non-negative MF (NMF)
   quiet: false            # no outputs to stdout
@@ -126,6 +126,71 @@ For one-class matrix factorization
 - `:one_class_col` - column-oriented pair-wise logarithmic loss
 - `:one_class_l2` - squared error (L2-norm)
 
+## Metrics
+
+Calculate RMSE (for real-valued MF)
+
+```ruby
+model.rmse(data)
+```
+
+Calculate MAE (for real-valued MF)
+
+```ruby
+model.mae(data)
+```
+
+Calculate generalized KL-divergence (for non-negative real-valued MF)
+
+```ruby
+model.gkl(data)
+```
+
+Calculate logarithmic loss (for binary MF)
+
+```ruby
+model.logloss(data)
+```
+
+Calculate accuracy (for binary MF)
+
+```ruby
+model.accuracy(data)
+```
+
+Calculate MPR (for one-class MF)
+
+```ruby
+model.mpr(data, transpose)
+```
+
+Calculate AUC (for one-class MF)
+
+```ruby
+model.auc(data, transpose)
+```
+
+## Example
+
+Download the [MovieLens 100K dataset](https://grouplens.org/datasets/movielens/100k/) and use:
+
+```ruby
+require "csv"
+
+train_set = Libmf::Matrix.new
+valid_set = Libmf::Matrix.new
+
+CSV.foreach("u.data", col_sep: "\t").with_index do |row, i|
+  data = i < 80000 ? train_set : valid_set
+  data.push(row[0].to_i, row[1].to_i, row[2].to_f)
+end
+
+model = Libmf::Model.new(factors: 20)
+model.fit(train_set, eval_set: valid_set)
+
+puts model.rmse(valid_set)
+```
+
 ## Performance
 
 For performance, read data directly from files
@@ -172,7 +237,7 @@ Everyone is encouraged to help improve this project. Here are a few ways you can
 To get started with development:
 
 ```sh
-git clone --recursive https://github.com/ankane/libmf-ruby.git
+git clone https://github.com/ankane/libmf-ruby.git
 cd libmf-ruby
 bundle install
 bundle exec rake vendor:all

data/lib/libmf/ffi.rb

@@ -59,7 +59,14 @@ module Libmf
     attach_function :mf_destroy_model, [:pointer], :void
     attach_function :mf_train, [Problem.by_ref, Parameter.by_value], Model.auto_ptr
     attach_function :mf_train_with_validation, [Problem.by_ref, Problem.by_ref, Parameter.by_value], Model.auto_ptr
-    attach_function :mf_predict, [Model.by_ref, :int, :int], :float
     attach_function :mf_cross_validation, [Problem.by_ref, :int, Parameter.by_value], :double
+    attach_function :mf_predict, [Model.by_ref, :int, :int], :float
+    attach_function :calc_rmse, [Problem.by_ref, Model.by_ref], :double
+    attach_function :calc_mae, [Problem.by_ref, Model.by_ref], :double
+    attach_function :calc_gkl, [Problem.by_ref, Model.by_ref], :double
+    attach_function :calc_logloss, [Problem.by_ref, Model.by_ref], :double
+    attach_function :calc_accuracy, [Problem.by_ref, Model.by_ref], :double
+    attach_function :calc_mpr, [Problem.by_ref, Model.by_ref, :bool], :double
+    attach_function :calc_auc, [Problem.by_ref, Model.by_ref, :bool], :double
   end
 end

data/lib/libmf/model.rb

@@ -35,6 +35,13 @@ module Libmf
       status = FFI.mf_save_model(model, path)
       raise Error, "Cannot save model" if status != 0
     end
+    alias_method :save, :save_model
+
+    def self.load(path)
+      model = Model.new
+      model.load_model(path)
+      model
+    end
 
     def load_model(path)
       @model = FFI.mf_load_model(path)
@@ -65,6 +72,34 @@ module Libmf
       _factors(model[:q], columns, format)
     end
 
+    def rmse(data)
+      FFI.calc_rmse(create_problem(data), model)
+    end
+
+    def mae(data)
+      FFI.calc_mae(create_problem(data), model)
+    end
+
+    def gkl(data)
+      FFI.calc_gkl(create_problem(data), model)
+    end
+
+    def logloss(data)
+      FFI.calc_logloss(create_problem(data), model)
+    end
+
+    def accuracy(data)
+      FFI.calc_accuracy(create_problem(data), model)
+    end
+
+    def mpr(data, transpose)
+      FFI.calc_mpr(create_problem(data), model, transpose)
+    end
+
+    def auc(data, transpose)
+      FFI.calc_auc(create_problem(data), model, transpose)
+    end
+
     private
 
     def _factors(ptr, n, format)
@@ -106,13 +141,13 @@ module Libmf
       options[:bins] ||= 25 unless options[:nr_bins]
       options[:copy_data] = false unless options.key?(:copy_data)
       options_map = {
-        :loss => :fun,
-        :factors => :k,
-        :threads => :nr_threads,
-        :bins => :nr_bins,
-        :iterations => :nr_iters,
-        :learning_rate => :eta,
-        :nmf => :do_nmf
+        loss: :fun,
+        factors: :k,
+        threads: :nr_threads,
+        bins: :nr_bins,
+        iterations: :nr_iters,
+        learning_rate: :eta,
+        nmf: :do_nmf
       }
       options.each do |k, v|
         k = options_map[k] if options_map[k]

data/lib/libmf/version.rb

@@ -1,3 +1,3 @@
 module Libmf
-  VERSION = "0.2.6"
+  VERSION = "0.4.0"
 end

data/lib/libmf.rb

@@ -2,9 +2,9 @@
 require "ffi"
 
 # modules
-require "libmf/matrix"
-require "libmf/model"
-require "libmf/version"
+require_relative "libmf/matrix"
+require_relative "libmf/model"
+require_relative "libmf/version"
 
 module Libmf
   class Error < StandardError; end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: libmf
 version: !ruby/object:Gem::Version
-  version: 0.2.6
+  version: 0.4.0
 platform: ruby
 authors:
 - Andrew Kane
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2021-12-02 00:00:00.000000000 Z
+date: 2024-10-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ffi
@@ -58,14 +58,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '2.4'
+      version: '3.1'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.2.32
+rubygems_version: 3.5.16
 signing_key:
 specification_version: 4
 summary: Large-scale sparse matrix factorization for Ruby