xgb 0.9.0 → 0.11.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ba232bc8a9c27bc9cfb2b7f87afc581acf9a5464df9fccf1b206d912704771fe
-  data.tar.gz: 608ca80f01e81d32a31b796364de14096c89556543fa2892efd86bbf86fac96a
+  metadata.gz: f5fd1670cc6fc20a352e85facd9abb1923195c14022e7f909fc588217e758973
+  data.tar.gz: 9bfe06f330635d6f4f5bbf71223be5a7d2f4891e394c12f697a10a18215bf694
 SHA512:
-  metadata.gz: af38ca270cf3d12a8a7757ff5f32a41f5d68e43f9ad08c585df8efae835d014ddcac51201019aa339143cc081c5c69ef5914d8de06cb4326ccb2864b5b04eba2
-  data.tar.gz: 787bddc1c867d648ffe13c783aede98c62917feca536d1f8fe27e7992e5f2945126ad15b1a4e3ff48cc041723d299cf3cbacdd5557083983575eae22a9583cd8
+  metadata.gz: 90fe5e7cc050bf8d8a2079974501cb50ac886bcb1482160afca50c1cf598d836b59c324630cf00236ac8e089fe57fb618335dde4c9a22dd3d4a5bc7fd031f41c
+  data.tar.gz: 4d87ee477501874f53633c286bec193da8f731f7e14f3f9008ed2da68fd7e3f3f594c25bc47d8f940813546208ff50981731ab67204d140c0a9cdebba6727565

data/CHANGELOG.md

@@ -1,3 +1,12 @@
+## 0.11.0 (2025-10-18)
+
+- Updated XGBoost to 3.1.0
+- Dropped support for Ruby < 3.2
+
+## 0.10.0 (2025-03-15)
+
+- Updated XGBoost to 3.0.0
+
 ## 0.9.0 (2024-10-17)
 
 - Updated XGBoost to 2.1.1

data/NOTICE.txt

@@ -1,5 +1,5 @@
-Copyright XGBoost contributors
-Copyright 2019-2024 Andrew Kane
+Copyright 2014-2025 XGBoost contributors
+Copyright 2019-2025 Andrew Kane
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.

data/README.md

@@ -126,7 +126,8 @@ model.feature_importances
 Early stopping
 
 ```ruby
-model.fit(x, y, eval_set: [[x_test, y_test]], early_stopping_rounds: 5)
+model = XGBoost::Regressor.new(early_stopping_rounds: 5)
+model.fit(x, y, eval_set: [[x_test, y_test]])
 ```
 
 ## Data

data/lib/xgboost/booster.rb

@@ -47,6 +47,11 @@ module XGBoost
       json_string.read_pointer.read_string(length.read_uint64).force_encoding(Encoding::UTF_8)
     end
 
+    def reset
+      check_call FFI.XGBoosterReset(handle)
+      self
+    end
+
     def attr(key)
       ret = ::FFI::MemoryPointer.new(:pointer)
       success = ::FFI::MemoryPointer.new(:int)
@@ -178,7 +183,7 @@ module XGBoost
 
       names = feature_names || []
       fnames = array_of_pointers(names.map { |fname| string_pointer(fname) })
-      ftypes = array_of_pointers(feature_types || Array.new(names.size, string_pointer("float")))
+      ftypes = array_of_pointers(feature_types&.map { |v| string_pointer(v) } || Array.new(names.size, string_pointer("float")))
 
       check_call FFI.XGBoosterDumpModelExWithFeatures(handle, names.size, fnames, ftypes, with_stats ? 1 : 0, dump_format, out_len, out_result)
 

data/lib/xgboost/callback_container.rb

@@ -110,7 +110,6 @@ module XGBoost
 
     def aggcv(rlist)
       cvmap = {}
-      idx = rlist[0].split[0]
       rlist.each do |line|
         arr = line.split
         arr[1..].each_with_index do |it, metric_idx|
@@ -118,7 +117,6 @@ module XGBoost
           (cvmap[[metric_idx, k]] ||= []) << v.to_f
         end
       end
-      msg = idx
       results = []
       cvmap.sort { |x| x[0][0] }.each do |(_, name), s|
         mean = mean(s)

data/lib/xgboost/classifier.rb

@@ -1,6 +1,6 @@
 module XGBoost
   class Classifier < Model
-    def initialize(n_estimators: 100, objective: "binary:logistic", importance_type: "gain", **options)
+    def initialize(objective: "binary:logistic", **options)
       super
     end
 
@@ -18,7 +18,7 @@ module XGBoost
 
       @booster = XGBoost.train(params, dtrain,
         num_boost_round: @n_estimators,
-        early_stopping_rounds: early_stopping_rounds,
+        early_stopping_rounds: early_stopping_rounds || @early_stopping_rounds,
         verbose_eval: verbose,
         evals: evals
       )

data/lib/xgboost/dmatrix.rb

@@ -34,6 +34,17 @@ module XGBoost
       elsif rover?(data)
         nrow, ncol = data.shape
         feature_names = data.keys
+        feature_types =
+          data.types.map do |_, v|
+            v = v.to_s
+            if v.start_with?("int") || v.start_with?("uint")
+              "int"
+            elsif v.start_with?("float")
+              "float"
+            else
+              raise Error, "Unknown feature type: #{v}"
+            end
+          end
         data = data.to_numo
       else
         nrow = data.count
@@ -56,7 +67,7 @@ module XGBoost
       check_call FFI.XGDMatrixCreateFromMat(c_data, nrow, ncol, missing, out)
       @handle = ::FFI::AutoPointer.new(out.read_pointer, FFI.method(:XGDMatrixFree))
 
-      self.feature_names = feature_names || ncol.times.map { |i| "f#{i}" }
+      self.feature_names = feature_names if feature_names
       self.feature_types = feature_types if feature_types
 
       self.label = label if label
@@ -76,9 +87,15 @@ module XGBoost
     end
 
     def group=(group)
-      c_data = ::FFI::MemoryPointer.new(:int, group.size)
-      c_data.write_array_of_int(group)
-      check_call FFI.XGDMatrixSetUIntInfo(handle, "group", c_data, group.size)
+      c_data = ::FFI::MemoryPointer.new(:uint32, group.size)
+      c_data.write_array_of_uint32(group)
+      interface = {
+        shape: [group.length],
+        typestr: "|u4",
+        data: [c_data.address, false],
+        version: 3
+      }
+      check_call FFI.XGDMatrixSetInfoFromInterface(handle, "group", JSON.generate(interface))
     end
 
     def label
@@ -89,6 +106,10 @@ module XGBoost
       float_info("weight")
     end
 
+    def group
+      uint_info("group_ptr")
+    end
+
     def num_row
       out = ::FFI::MemoryPointer.new(:uint64)
       check_call FFI.XGDMatrixNumRow(handle, out)
@@ -234,7 +255,15 @@ module XGBoost
       out_len = ::FFI::MemoryPointer.new(:uint64)
       out_dptr = ::FFI::MemoryPointer.new(:float, num_row)
       check_call FFI.XGDMatrixGetFloatInfo(handle, field, out_len, out_dptr)
-      out_dptr.read_pointer.read_array_of_float(num_row)
+      out_dptr.read_pointer.null? ? nil : out_dptr.read_pointer.read_array_of_float(num_row)
+    end
+
+    def uint_info(field)
+      num_row ||= num_row()
+      out_len = ::FFI::MemoryPointer.new(:uint64)
+      out_dptr = ::FFI::MemoryPointer.new(:uint32, num_row)
+      check_call FFI.XGDMatrixGetUIntInfo(handle, field, out_len, out_dptr)
+      out_dptr.read_pointer.null? ? nil : out_dptr.read_pointer.read_array_of_uint32(num_row)
     end
 
     def validate_feature_info(feature_info, n_features, is_column_split, name)

data/lib/xgboost/ffi.rb

@@ -21,7 +21,7 @@ module XGBoost
 
     # dmatrix
     attach_function :XGDMatrixCreateFromMat, %i[pointer uint64 uint64 float pointer], :int
-    attach_function :XGDMatrixSetUIntInfo, %i[pointer string pointer uint64], :int
+    attach_function :XGDMatrixSetInfoFromInterface, %i[pointer string string], :int
     attach_function :XGDMatrixSetStrFeatureInfo, %i[pointer string pointer uint64], :int
     attach_function :XGDMatrixGetStrFeatureInfo, %i[pointer string pointer pointer], :int
     attach_function :XGDMatrixNumRow, %i[pointer pointer], :int
@@ -33,12 +33,14 @@ module XGBoost
     attach_function :XGDMatrixSaveBinary, %i[pointer string int], :int
     attach_function :XGDMatrixSetFloatInfo, %i[pointer string pointer uint64], :int
     attach_function :XGDMatrixGetFloatInfo, %i[pointer string pointer pointer], :int
+    attach_function :XGDMatrixGetUIntInfo, %i[pointer string pointer pointer], :int
 
     # booster
     attach_function :XGBoosterCreate, %i[pointer int pointer], :int
     attach_function :XGBoosterUpdateOneIter, %i[pointer int pointer], :int
     attach_function :XGBoosterEvalOneIter, %i[pointer int pointer pointer uint64 pointer], :int
     attach_function :XGBoosterFree, %i[pointer], :int
+    attach_function :XGBoosterReset, %i[pointer], :int
     attach_function :XGBoosterBoostedRounds, %i[pointer pointer], :int
     attach_function :XGBoosterSetParam, %i[pointer string string], :int
     attach_function :XGBoosterGetNumFeature, %i[pointer pointer], :int

data/lib/xgboost/model.rb

@@ -2,10 +2,11 @@ module XGBoost
   class Model
     attr_reader :booster
 
-    def initialize(n_estimators: 100, importance_type: "gain", **options)
+    def initialize(n_estimators: 100, importance_type: "gain", early_stopping_rounds: nil, **options)
       @params = options
       @n_estimators = n_estimators
       @importance_type = importance_type
+      @early_stopping_rounds = early_stopping_rounds
     end
 
     def predict(data)
@@ -18,12 +19,13 @@ module XGBoost
     end
 
     def load_model(fname)
-      @booster = Booster.new(params: @params, model_file: fname)
+      @booster = Booster.new(model_file: fname)
     end
 
     def feature_importances
       score = @booster.score(importance_type: @importance_type)
-      scores = @booster.feature_names.map { |k| score[k] || 0.0 }
+      feature_names = @booster.feature_names || @booster.num_features.times.map { |i| "f#{i}" }
+      scores = feature_names.map { |k| score[k] || 0.0 }
       total = scores.sum.to_f
       scores.map { |s| s / total }
     end

data/lib/xgboost/regressor.rb

@@ -1,6 +1,6 @@
 module XGBoost
   class Regressor < Model
-    def initialize(n_estimators: 100, objective: "reg:squarederror", importance_type: "gain", **options)
+    def initialize(objective: "reg:squarederror", **options)
       super
     end
 
@@ -10,7 +10,7 @@ module XGBoost
 
       @booster = XGBoost.train(@params, dtrain,
         num_boost_round: @n_estimators,
-        early_stopping_rounds: early_stopping_rounds,
+        early_stopping_rounds: early_stopping_rounds || @early_stopping_rounds,
         verbose_eval: verbose,
         evals: evals
       )

data/lib/xgboost/utils.rb

@@ -23,11 +23,9 @@ module XGBoost
     end
 
     def from_cstr_to_rbstr(data, length)
-      res = []
-      length.read_uint64.times do |i|
-        res << data.read_pointer[i * ::FFI::Pointer.size].read_pointer.read_string.force_encoding(Encoding::UTF_8)
+      data.read_pointer.read_array_of_pointer(length.read_uint64).map do |ptr|
+        ptr.read_string.force_encoding(Encoding::UTF_8)
       end
-      res
     end
   end
 end

data/lib/xgboost/version.rb

@@ -1,3 +1,3 @@
 module XGBoost
-  VERSION = "0.9.0"
+  VERSION = "0.11.0"
 end

data/lib/xgboost.rb

@@ -91,7 +91,7 @@ module XGBoost
         evals_result.merge!(cb_container.history)
       end
 
-      bst
+      bst.reset
     end
 
     def cv(

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: xgb
 version: !ruby/object:Gem::Version
-  version: 0.9.0
+  version: 0.11.0
 platform: ruby
 authors:
 - Andrew Kane
-autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-10-17 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ffi
@@ -24,7 +23,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-description:
 email: andrew@ankane.org
 executables: []
 extensions: []
@@ -73,7 +71,6 @@ homepage: https://github.com/ankane/xgboost-ruby
 licenses:
 - Apache-2.0
 metadata: {}
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -81,15 +78,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '3.1'
+      version: '3.2'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.5.16
-signing_key:
+rubygems_version: 3.6.9
 specification_version: 4
 summary: High performance gradient boosting for Ruby
 test_files: []