numo-libsvm 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: abc7fcd5ce86d8e2d0576290a8053b77f4e06d5e
+  data.tar.gz: f731f73a38b034e36f9b79164c5d0cc27ca30f1a
+SHA512:
+  metadata.gz: 82c172d2cd872c3cae5bcbc685d175489401ca19b963af80aac4c8f380fefe40d2c54cb045d471ff1887a792063c694c222781cb1ca7d34082e45f3726d2d848
+  data.tar.gz: 6f483ae0c4c8ed902bfa75727dbeaa15559f5dca5c0903d624e93cc6a7239e635441394e9a325b752ac2b0e6894d149e3996e15c6d1e0a7935e08c3aed613a91

data/.gitignore

@@ -0,0 +1,20 @@
+/.bundle/
+/.yardoc
+/Gemfile.lock
+/_yardoc/
+/coverage/
+/doc/
+/pkg/
+/spec/reports/
+/tmp/
+/bin/
+
+# rspec failure tracking
+.rspec_status
+
+*.swp
+*.bundle
+mkmf.log
+tags
+.DS_Store
+.ruby-version

data/.rspec

@@ -0,0 +1,3 @@
+--format documentation
+--color
+--require spec_helper

data/.travis.yml

@@ -0,0 +1,14 @@
+---
+sudo: true
+os: linux
+dist: bionic
+language: ruby
+cache: bundler
+rvm:
+  - '2.4'
+  - '2.5'
+  - '2.6'
+
+before_install:
+  - sudo apt-get install -y libsvm-dev
+  - gem install bundler -v 2.0.2

data/CHANGELOG.md

@@ -0,0 +1,2 @@
+# 0.1.0
+- First release.

data/CODE_OF_CONDUCT.md

@@ -0,0 +1,74 @@
+# Contributor Covenant Code of Conduct
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as
+contributors and maintainers pledge to making participation in our project and
+our community a harassment-free experience for everyone, regardless of age, body
+size, disability, ethnicity, gender identity and expression, level of experience,
+nationality, personal appearance, race, religion, or sexual identity and
+orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment
+include:
+
+* Using welcoming and inclusive language
+* Being respectful of differing viewpoints and experiences
+* Gracefully accepting constructive criticism
+* Focusing on what is best for the community
+* Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery and unwelcome sexual attention or
+advances
+* Trolling, insulting/derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or electronic
+  address, without explicit permission
+* Other conduct which could reasonably be considered inappropriate in a
+  professional setting
+
+## Our Responsibilities
+
+Project maintainers are responsible for clarifying the standards of acceptable
+behavior and are expected to take appropriate and fair corrective action in
+response to any instances of unacceptable behavior.
+
+Project maintainers have the right and responsibility to remove, edit, or
+reject comments, commits, code, wiki edits, issues, and other contributions
+that are not aligned to this Code of Conduct, or to ban temporarily or
+permanently any contributor for other behaviors that they deem inappropriate,
+threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies both within project spaces and in public spaces
+when an individual is representing the project or its community. Examples of
+representing a project or community include using an official project e-mail
+address, posting via an official social media account, or acting as an appointed
+representative at an online or offline event. Representation of a project may be
+further defined and clarified by project maintainers.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be
+reported by contacting the project team at yoshoku@outlook.com. All
+complaints will be reviewed and investigated and will result in a response that
+is deemed necessary and appropriate to the circumstances. The project team is
+obligated to maintain confidentiality with regard to the reporter of an incident.
+Further details of specific enforcement policies may be posted separately.
+
+Project maintainers who do not follow or enforce the Code of Conduct in good
+faith may face temporary or permanent repercussions as determined by other
+members of the project's leadership.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
+available at [http://contributor-covenant.org/version/1/4][version]
+
+[homepage]: http://contributor-covenant.org
+[version]: http://contributor-covenant.org/version/1/4/

data/Gemfile

@@ -0,0 +1,4 @@
+source "https://rubygems.org"
+
+# Specify your gem's dependencies in numo-libsvm.gemspec
+gemspec

data/LICENSE.txt

@@ -0,0 +1,27 @@
+Copyright (c) 2019 Atsushi Tatsuma
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* 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.
+
+* 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.
+
+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.

data/README.md

@@ -0,0 +1,180 @@
+# Numo::Libsvm
+
+[![Build Status](https://travis-ci.org/yoshoku/numo-libsvm.svg?branch=master)](https://travis-ci.org/yoshoku/numo-libsvm)
+[![Gem Version](https://badge.fury.io/rb/numo-libsvm.svg)](https://badge.fury.io/rb/numo-libsvm)
+[![BSD 3-Clause License](https://img.shields.io/badge/License-BSD%203--Clause-blue.svg)](https://github.com/yoshoku/numo-libsvm/blob/master/LICENSE.txt)
+
+Numo::Libsvm is a Ruby gem binding to the [LIBSVM](https://github.com/cjlin1/libsvm) library.
+LIBSVM is one of the famous libraries that implemented Support Vector Machines,
+and provides functions for support vector classifier, regression, and distribution estimation.
+Numo::Libsvm makes to use the LIBSVM functions with dataset represented by [Numo::NArray](https://github.com/ruby-numo/numo-narray).
+
+Note: There are other useful Ruby gems binding to LIBSVM:
+[rb-libsvm](https://github.com/febeling/rb-libsvm) by C. Florian Ebeling,
+[libsvm-ruby-swig](https://github.com/tomz/libsvm-ruby-swig) by Tom Zeng,
+and [jrb-libsvm](https://github.com/andreaseger/jrb-libsvm) by Andreas Eger.
+
+## Installation
+Numo::Libsvm does not bundle LIBSVM unlike rb-libsvm. You need to install LIBSVM in advance along your environment.
+
+macOS:
+
+    $ brew install libsvm
+
+Ubuntu:
+
+    $ sudo apt-get install libsvm-dev
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'numo-libsvm'
+```
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install numo-libsvm
+
+## Usage
+
+### Preparation
+
+In the following examples, we use [red-datasets](https://github.com/red-data-tools/red-datasets) to download dataset.
+
+    $ gem install red-datasets-numo-narray
+
+### Example 1. Cross-validation
+
+We conduct cross validation of support vector classifier on [Iris dataset](https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#iris).
+
+```ruby
+require 'numo/narray'
+require 'numo/libsvm'
+require 'datasets-numo-narray'
+
+# Download Iris dataset.
+puts 'Download dataset.'
+iris = Datasets::LIBSVM.new('iris').to_narray
+x = iris[true, 1..-1]
+y = iris[true, 0]
+
+# Define parameters of C-SVC with RBF Kernel.
+param = {
+  svm_type: Numo::Libsvm::SvmType::C_SVC,
+  kernel_type: Numo::Libsvm::KernelType::RBF,
+  gamma: 1.0,
+  C: 1
+}
+
+# Perform 5-cross validation.
+puts 'Perform cross validation.'
+n_folds = 5
+predicted = Numo::Libsvm.cv(x, y, param, n_folds)
+
+# Print mean accuracy.
+mean_accuracy = y.eq(predicted).count.fdiv(y.size)
+puts "Accuracy: %.1f %%" % (100 * mean_accuracy)
+```
+
+Execution result in the following:
+
+```sh
+Download dataset.
+Perform cross validation.
+Accuracy: 96.0 %
+```
+
+### Example 2. Pendigits dataset classification
+
+We first train the support vector classifier with RBF kernel using training [pendigits dataset](https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#pendigits).
+
+```ruby
+require 'numo/narray'
+require 'numo/libsvm'
+require 'datasets-numo-narray'
+
+# Download pendigits training dataset.
+puts 'Download dataset.'
+pendigits = Datasets::LIBSVM.new('pendigits').to_narray
+x = pendigits[true, 1..-1]
+y = pendigits[true, 0]
+
+# Define parameters of C-SVC with RBF Kernel.
+param = {
+  svm_type: Numo::Libsvm::SvmType::C_SVC,
+  kernel_type: Numo::Libsvm::KernelType::RBF,
+  gamma: 0.0001,
+  C: 10,
+  shrinking: true
+}
+
+# Perform training procedure.
+puts 'Train support vector machine.'
+model = Numo::Libsvm.train(x, y, param)
+
+# Save parameters and trained model.
+puts 'Save parameters and model with Marshal.'
+File.open('pendigits.dat', 'wb') { |f| f.write(Marshal.dump([param, model])) }
+```
+
+```sh
+$ ruby train.rb
+Download dataset.
+Train support vector machine.
+Save paramters and model with Marshal.
+```
+
+We then predict labels of testing dataset, and evaluate the classifier.
+
+```ruby
+require 'numo/narray'
+require 'numo/libsvm'
+require 'datasets-numo-narray'
+
+# Download pendigits testing dataset.
+puts 'Download dataset.'
+pendigits_test = Datasets::LIBSVM.new('pendigits', note: 'testing').to_narray
+x = pendigits_test[true, 1..-1]
+y = pendigits_test[true, 0]
+
+# Load parameter and model.
+puts 'Load parameter and model.'
+param, model = Marshal.load(File.binread('pendigits.dat'))
+
+# Predict labels.
+puts 'Predict labels.'
+predicted = Numo::Libsvm.predict(x, param, model)
+
+# Evaluate classification results.
+mean_accuracy = y.eq(predicted).count.fdiv(y.size)
+puts "Accuracy: %.1f %%" % (100 * mean_accuracy)
+```
+
+```sh
+$ ruby test.rb
+Download dataset.
+Load parameter and model.
+Predict labels.
+Accuracy: 98.3 %
+```
+
+### Note
+The hyperparameter of SVM is given with Ruby Hash on Numo::Libsvm.
+The hash key of hyperparameter and its meaning match the struct svm_parameter of LIBSVM.
+The svm_parameter is detailed in [LIBSVM README](https://github.com/cjlin1/libsvm/blob/master/README).
+
+## Contributing
+
+Bug reports and pull requests are welcome on GitHub at https://github.com/yoshoku/numo-libsvm. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [Contributor Covenant](http://contributor-covenant.org) code of conduct.
+
+## License
+
+The gem is available as open source under the terms of the [BSD-3-Clause License](https://opensource.org/licenses/BSD-3-Clause).
+
+## Code of Conduct
+
+Everyone interacting in the Numo::Libsvm project’s codebases, issue trackers, chat rooms and mailing lists is expected to follow the [code of conduct](https://github.com/[USERNAME]/numo-libsvm/blob/master/CODE_OF_CONDUCT.md).

data/Rakefile

@@ -0,0 +1,15 @@
+require 'bundler/gem_tasks'
+require 'rspec/core/rake_task'
+
+RSpec::Core::RakeTask.new(:spec)
+
+require 'rake/extensiontask'
+
+task :build => :compile
+
+Rake::ExtensionTask.new('libsvmext') do |ext|
+  ext.ext_dir = 'ext/numo/libsvm'
+  ext.lib_dir = 'lib/numo/libsvm'
+end
+
+task :default => [:clobber, :compile, :spec]

data/ext/numo/libsvm/converter.c

@@ -0,0 +1,162 @@
+
+#include "converter.h"
+
+VALUE int_vec_to_nary(int* const arr, int const size)
+{
+  int i;
+  size_t shape[1] = { size };
+  VALUE v = rb_narray_new(numo_cInt32, 1, shape);
+  int32_t* vp = (int32_t*)na_get_pointer_for_write(v);
+  for (i = 0; i < size; i++) { vp[i] = (int32_t)arr[i]; }
+  return v;
+}
+
+int* nary_to_int_vec(VALUE vec_val)
+{
+  int i;
+  int n_elements;
+  narray_t* vec_nary;
+  int32_t* vec_pt;
+  int* vec;
+
+  if (vec_val == Qnil) return NULL;
+
+  GetNArray(vec_val, vec_nary);
+  n_elements = (int)NA_SHAPE(vec_nary)[0];
+
+  vec = ALLOC_N(int, n_elements);
+  vec_pt = (int32_t*)na_get_pointer_for_read(vec_val);
+  for (i = 0; i < n_elements; i++) { vec[i] = (int)vec_pt[i]; }
+
+  return vec;
+}
+
+VALUE dbl_vec_to_nary(double* const arr, int const size)
+{
+  int i;
+  size_t shape[1] = { size };
+  VALUE v = rb_narray_new(numo_cDFloat, 1, shape);
+  double* vp = (double*)na_get_pointer_for_write(v);
+  for (i = 0; i < size; i++) { vp[i] = arr[i]; }
+  return v;
+}
+
+double* nary_to_dbl_vec(VALUE vec_val)
+{
+  int n_elements;
+  narray_t* vec_nary;
+  double* vec_pt;
+  double* vec;
+
+  if (vec_val == Qnil) return NULL;
+
+  GetNArray(vec_val, vec_nary);
+  n_elements = (int)NA_SHAPE(vec_nary)[0];
+
+  vec = ALLOC_N(double, n_elements);
+  vec_pt = (double*)na_get_pointer_for_read(vec_val);
+  memcpy(vec, vec_pt, n_elements * sizeof(double));
+
+  return vec;
+}
+
+VALUE dbl_mat_to_nary(double** const mat, int const n_rows, int const n_cols)
+{
+  int i, j;
+  size_t shape[2] = { n_rows, n_cols };
+  VALUE v = rb_narray_new(numo_cDFloat, 2, shape);
+  double* vp = (double*)na_get_pointer_for_write(v);
+
+  for (i = 0; i < n_rows; i++) {
+    for (j = 0; j < n_cols; j++) {
+      vp[i * n_cols + j] = mat[i][j];
+    }
+  }
+
+  return v;
+}
+
+double** nary_to_dbl_mat(VALUE mat_val)
+{
+  int i, j;
+  int n_rows, n_cols;
+  narray_t* mat_nary;
+  double* mat_pt;
+  double** mat;
+
+  if (mat_val == Qnil) return NULL;
+
+  GetNArray(mat_val, mat_nary);
+  n_rows = (int)NA_SHAPE(mat_nary)[0];
+  n_cols = (int)NA_SHAPE(mat_nary)[1];
+
+  mat_pt = (double*)na_get_pointer_for_read(mat_val);
+  mat = ALLOC_N(double*, n_rows);
+  for (i = 0; i < n_rows; i++) {
+    mat[i] = ALLOC_N(double, n_cols);
+    for (j = 0; j < n_cols; j++) {
+      mat[i][j] = mat_pt[i * n_cols + j];
+    }
+  }
+
+  return mat;
+}
+
+VALUE svm_nodes_to_nary(struct svm_node** const support_vecs, const int n_support_vecs)
+{
+  int i, j;
+  int n_dimensions = 0;
+  size_t shape[2] = { n_support_vecs, 1 };
+  VALUE v;
+  double* vp;
+
+  for (i = 0; i < n_support_vecs; i++) {
+    for (j = 0; support_vecs[i][j].index != -1; j++) {
+      if (n_dimensions < support_vecs[i][j].index) {
+        n_dimensions = support_vecs[i][j].index;
+      }
+    }
+  }
+
+  shape[1] = n_dimensions;
+  v = rb_narray_new(numo_cDFloat, 2, shape);
+  vp = (double*)na_get_pointer_for_write(v);
+  memset(vp, 0, n_support_vecs * n_dimensions * sizeof(double));
+
+  for (i = 0; i < n_support_vecs; i++) {
+    for (j = 0; support_vecs[i][j].index != -1; j++) {
+      vp[i * n_dimensions + support_vecs[i][j].index - 1] = support_vecs[i][j].value;
+    }
+  }
+
+  return v;
+}
+
+struct svm_node** nary_to_svm_nodes(VALUE model_val)
+{
+  int i, j;
+  int n_rows, n_cols;
+  narray_t* model_nary;
+  double* model_pt;
+  struct svm_node** support_vecs;
+
+  if (model_val == Qnil) return NULL;
+
+  GetNArray(model_val, model_nary);
+  n_rows = (int)NA_SHAPE(model_nary)[0];
+  n_cols = (int)NA_SHAPE(model_nary)[1];
+
+  model_pt = (double*)na_get_pointer_for_read(model_val);
+  support_vecs = ALLOC_N(struct svm_node*, n_rows);
+  for (i = 0; i < n_rows; i++) {
+    support_vecs[i] = ALLOC_N(struct svm_node, n_cols + 1);
+    for (j = 0; j < n_cols; j++) {
+      support_vecs[i][j].index = j + 1;
+      support_vecs[i][j].value = model_pt[i * n_cols + j];
+    }
+    support_vecs[i][n_cols].index = -1;
+    support_vecs[i][n_cols].value = 0.0;
+  }
+
+  return support_vecs;
+}