scs 0.3.2 → 0.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: db74c6ffd6696e799a18d2f5d02f5b753020a2535e539cdd26abac388f1ea046
-  data.tar.gz: f7d5b3b98b58d812fb923193303b7f0bfcf7cb2972de2ae34a392fb9f4a5bd40
+  metadata.gz: 9005db60b3ebc21bfe74e18fbd90db752801cce6232c5dcbbf9746ed6a3772b9
+  data.tar.gz: 8d1eade7f1501cbf46e820a2e7b20133f008e14d2bcc098629643a9b750b55b3
 SHA512:
-  metadata.gz: cf7a4948ba7a21865142a421da3e683904951faf7e602d429722cf05bdbae559dcc15484e6cb4afa6bf683ecf681757370bfa3fa7b01dab6620b300462eea703
-  data.tar.gz: 493da1d4fbe2fdb7180fffad3599137500d437b082d788365bf828dd8845469b3bf0d63863ea467c013cc2f874c5b04ab935c7129d45b1c890278dda7ad21039
+  metadata.gz: 98ecf4f8beb7215e0ca44a57e3345b257054948c311e8e749f31d698e27c7f6c66ccbaaa71bbb091a92d4a23809b9b98a5d65767ee8d58e78cc7a8f08a32e042
+  data.tar.gz: 4701d88bad2ae68509fba757a922019400a78f8d8c2bdf0e33f35f7e409c49b0e8306d0cc3281a542508e3a73176042bf16a6039aee5101430ff378689791686

data/CHANGELOG.md

@@ -1,3 +1,13 @@
+## 0.4.1 (2023-04-09)
+
+- Updated SCS to 3.2.3
+
+## 0.4.0 (2022-07-06)
+
+- Updated SCS to 3.2.0
+- Added `Matrix` class
+- Dropped support for Ruby < 2.7
+
 ## 0.3.2 (2022-01-17)
 
 - Updated SCS to 3.1.1

data/LICENSE.txt

@@ -1,7 +1,7 @@
 The MIT License (MIT)
 
 Copyright (c) 2012 Brendan O'Donoghue (bodonoghue85@gmail.com)
-Copyright (c) 2019-2021 Andrew Kane
+Copyright (c) 2019-2023 Andrew Kane
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

data/README.md

@@ -2,7 +2,7 @@
 
 [SCS](https://github.com/cvxgrp/scs) - the splitting conic solver - for Ruby
 
-:fire: Supports many different [problem types](https://www.cvxpy.org/tutorial/advanced/index.html#choosing-a-solver)
+Check out [Opt](https://github.com/ankane/opt) for a high-level interface
 
 [![Build Status](https://github.com/ankane/scs-ruby/workflows/build/badge.svg?branch=master)](https://github.com/ankane/scs-ruby/actions)
 
@@ -18,11 +18,16 @@ If installation fails, you may need to install [dependencies](#dependencies).
 
 ## Getting Started
 
-Prep the problem
+Prep the problem, like [this one](https://www.cvxgrp.org/scs/examples/python/basic_qp.html)
 
 ```ruby
-data = {a: [[1], [-1]], b: [1, 0], c: [-1]}
-cone = {q: [], l: 2}
+data = {
+  p: SCS::Matrix.from_dense([[3, -1], [-1, 2]]),
+  a: SCS::Matrix.from_dense([[-1, 1], [1, 0], [0, 1]]),
+  b: [-1, 0.3, -0.5],
+  c: [-1, -1]
+}
+cone = {z: 1, l: 2}
 ```
 
 And solve it
@@ -32,12 +37,36 @@ solver = SCS::Solver.new
 solver.solve(data, cone)
 ```
 
+## Data
+
+Matrices can be a sparse matrix
+
+```ruby
+a = SCS::Matrix.new(3, 2)
+a[0, 0] = 1
+a[1, 0] = 2
+# or
+SCS::Matrix.from_dense([[1, 0], [2, 0], [0, 0]])
+```
+
+Arrays can be Ruby arrays
+
+```ruby
+[1, 2, 3]
+```
+
+Or Numo arrays
+
+```ruby
+Numo::NArray.cast([1, 2, 3])
+```
+
 ## Settings
 
 Default values shown
 
 ```ruby
-solver.solve(data, cone, {
+solver.solve(data, cone,
   normalize: true,            # heuristic data rescaling
   scale: 0.1,                 # if normalized, rescales by this factor
   adaptive_scale: true,       # heuristically adapt dual scale through the solve
@@ -54,7 +83,7 @@ solver.solve(data, cone, {
   acceleration_interval: 10,  # iterations to run Anderson acceleration
   write_data_filename: nil,   # filename to write data if set
   log_csv_filename: nil       # write csv logs of various quantities
-})
+)
 ```
 
 ## Direct vs Indirect

data/lib/scs/matrix.rb

@@ -0,0 +1,72 @@
+module SCS
+  class Matrix
+    attr_reader :m, :n
+
+    def initialize(m, n)
+      @m = m
+      @n = n
+      @data = {}
+    end
+
+    def []=(row_index, column_index, value)
+      raise IndexError, "row index out of bounds" if row_index < 0 || row_index >= @m
+      raise IndexError, "column index out of bounds" if column_index < 0 || column_index >= @n
+      # dictionary of keys, optimized for converting to CSC
+      # TODO try COO for performance
+      if value == 0
+        (@data[column_index] ||= {}).delete(row_index)
+      else
+        (@data[column_index] ||= {})[row_index] = value
+      end
+    end
+
+    def to_csc
+      cx = []
+      ci = []
+      cp = []
+
+      # CSC format
+      # https://www.gormanalysis.com/blog/sparse-matrix-storage-formats/
+      cp << 0
+      n.times do |j|
+        (@data[j] || {}).sort_by { |k, v| k }.each do |k, v|
+          cx << v
+          ci << k
+        end
+        # cumulative column values
+        cp << cx.size
+      end
+
+      {
+        start: cp,
+        index: ci,
+        value: cx
+      }
+    end
+
+    # private, for tests
+    def nnz
+      @data.sum { |_, v| v.count }
+    end
+
+    def initialize_copy(other)
+      super
+      @data = @data.transform_values(&:dup)
+    end
+
+    def self.from_dense(data)
+      data = data.to_a
+      m = data.size
+      n = m > 0 ? data.first.size : 0
+
+      mtx = Matrix.new(m, n)
+      data.each_with_index do |row, i|
+        raise ArgumentError, "row has different number of columns" if row.size != n
+        row.each_with_index do |v, j|
+          mtx[i, j] = v if v != 0
+        end
+      end
+      mtx
+    end
+  end
+end

data/lib/scs/solver.rb

@@ -73,42 +73,35 @@ module SCS
       char_ptr[0, idx].map(&:chr).join
     end
 
-    # TODO add support sparse matrices
     def csc_matrix(mtx, upper: false)
-      mtx = mtx.to_a
-
-      m, n = shape(mtx)
-
-      cx = []
-      ci = []
-      cp = []
-
-      # CSC format
-      # https://www.gormanalysis.com/blog/sparse-matrix-storage-formats/
-      cp << 0
-      n.times do |j|
-        mtx.each_with_index do |row, i|
-          if row[j] != 0 && (!upper || i <= j)
-            cx << row[j]
-            ci << i
+      mtx = Matrix.from_dense(mtx) unless mtx.is_a?(Matrix)
+
+      if upper
+        # TODO improve performance
+        mtx = mtx.dup
+        mtx.m.times do |i|
+          mtx.n.times do |j|
+            mtx[i, j] = 0 if i > j
           end
         end
-        # cumulative column values
-        cp << cx.size
       end
 
+      csc = mtx.to_csc
+
       # construct matrix
       matrix = ffi::Matrix.malloc
-      matrix.x = float_array(cx)
-      matrix.i = int_array(ci)
-      matrix.p = int_array(cp)
-      matrix.m = m
-      matrix.n = n
+      matrix.x = float_array(csc[:value])
+      matrix.i = int_array(csc[:index])
+      matrix.p = int_array(csc[:start])
+      matrix.m = mtx.m
+      matrix.n = mtx.n
       matrix
     end
 
     def shape(a)
-      if defined?(Matrix) && a.is_a?(Matrix)
+      if a.is_a?(Matrix)
+        [a.m, a.n]
+      elsif defined?(::Matrix) && a.is_a?(::Matrix)
         [a.row_count, a.column_count]
       elsif defined?(Numo::NArray) && a.is_a?(Numo::NArray)
         a.shape
@@ -126,7 +119,7 @@ module SCS
 
       if data[:p]
         raise ArgumentError, "Bad p shape" if shape(data[:p]) != [n, n]
-        cdata.p = csc_matrix(data[:p])
+        cdata.p = csc_matrix(data[:p], upper: true)
       end
 
       raise ArgumentError, "Bad b size" if data[:b].to_a.size != m

data/lib/scs/version.rb

@@ -1,3 +1,3 @@
 module SCS
-  VERSION = "0.3.2"
+  VERSION = "0.4.1"
 end

data/lib/scs.rb

@@ -2,6 +2,7 @@
 require "fiddle/import"
 
 # modules
+require "scs/matrix"
 require "scs/solver"
 require "scs/version"
 

data/vendor/scs/CITATION.cff

@@ -10,8 +10,8 @@ authors:
 - family-names: "Boyd"
   given-names: "Stephen"
 title: "SCS: Spltting Conic Solver"
-version: 3.1.1
-date-released: 2021
+version: 3.2.3
+date-released: 2023
 url: "https://github.com/cvxgrp/scs"
 
 # Original SCS paper: