red-datasets 0.0.4 → 0.0.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 56c3132aec99f8dab763d68d0b731df0e5b2c97ea9f88a847021210fda94b9ac
-  data.tar.gz: 20e11d381718b801dac9c155385837b770c5067646ac235ade9c0f06f4d23cb8
+  metadata.gz: 1b52a97ab0ea10ea2d5ca5b873adab2e57aa8f8a71093e9e94c13da55bc8f774
+  data.tar.gz: 5f5525fc4cda3a9a57c08f6855fedd4da3c4be8b72fb7b35cee0e99133492c1f
 SHA512:
-  metadata.gz: 1c3ccf7b09e6064cacc47c04dd972ef4ba9318278630446ae60911412e9dc5dced5c988423b78a1c0ae68d71ede6136ebd2ee9c86f4247fd56bbf940f7997f7e
-  data.tar.gz: bebe49f31022dfeab1eb7626ef1eaf722d5dedf373dab89ba79e3e2afb4849ace50b5098688e4d0480f22719a758a086c288c600cdcb358099a7aa9d1e60eab1
+  metadata.gz: ae45fd3d9a6128ddca38c2b37a3a0c8fa89c831bdb1e14c1fdda2183be29385f74df6259ec4ae36934b8c0db57c9d2434208beeeaed854ea254ec6c327f21d64
+  data.tar.gz: 45209c6e106d78d008e73ab4414a933bc8f7333addf107c773b78f74359ef1fec262d5a22eeb9818db1512e1d0aab30f1baa00944bdd49eb7d8717f02454064b

data/README.md

@@ -48,11 +48,10 @@ cifar = Datasets::CIFAR.new(n_classes: 10, type: :train)
 cifar.metadata
 #=> #<struct Datasets::Metadata name="CIFAR-10", url="https://www.cs.toronto.edu/~kriz/cifar.html", licenses=nil, description="CIFAR-10 is 32x32 image dataset">licenses=nil, description="CIFAR-10 is 32x32 image datasets">
 cifar.each do |record|
-    p record.pixels
-    # => [59, 43, 50, 68, 98, 119, 139, 145, 149, 143, .....]
-    p record.label
-    # => 6
-  end
+  p record.pixels
+  # => [59, 43, 50, 68, 98, 119, 139, 145, 149, 143, .....]
+  p record.label
+  # => 6
 end
 ```
 
@@ -65,13 +64,12 @@ cifar = Datasets::CIFAR.new(n_classes: 100, type: :test)
 cifar.metadata
 #=> #<struct Datasets::Metadata name="CIFAR-100", url="https://www.cs.toronto.edu/~kriz/cifar.html", licenses=nil, description="CIFAR-100 is 32x32 image dataset">
 cifar.each do |record|
-    p record.pixels
-    #=> [199, 196, 195, 195, 196, 197, 198, 198, 199, .....]
-    p record.coarse_label
-    #=> 10
-    p record.fine_label
-    #=> 49
-  end
+  p record.pixels
+  #=> [199, 196, 195, 195, 196, 197, 198, 198, 199, .....]
+  p record.coarse_label
+  #=> 10
+  p record.fine_label
+  #=> 49
 end
 ```
 

data/doc/text/news.md

@@ -1,5 +1,26 @@
 # News
 
+## 0.0.5 - 2018-06-06
+
+### Improvements
+
+  * `Datasets::Table#[]`: Added.
+
+  * `Datasets::Table#fetch_values`: Added.
+
+  * `Datasets::Table#each`: Added.
+
+  * `Datasets::CIFAR`: Added pixels data to `Datasets::Table`.
+
+### Fixes
+
+  * Fixed indent of sample codes in README.
+    [GitHub#11][Patch by FURUSAWA Tomohiro]
+
+### Thanks
+
+  * FURUSAWA Tomohiro
+
 ## 0.0.4 - 2018-05-03
 
 ### Improvements

data/lib/datasets/cifar.rb

@@ -5,16 +5,24 @@ require_relative "dataset"
 
 module Datasets
   class CIFAR < Dataset
-    class Record10 < Struct.new(:data, :label)
+    module Pixelable
       def pixels
         data.unpack("C*")
       end
+
+      def to_h
+        hash = super
+        hash[:pixels] = pixels
+        hash
+      end
+    end
+
+    class Record10 < Struct.new(:data, :label)
+      include Pixelable
     end
 
     class Record100 < Struct.new(:data, :coarse_label, :fine_label)
-      def pixels
-        data.unpack("C*")
-      end
+      include Pixelable
     end
 
     def initialize(n_classes: 10, type: :train)

data/lib/datasets/table.rb

@@ -1,9 +1,31 @@
 module Datasets
   class Table
+    include Enumerable
+
     def initialize(dataset)
       @dataset = dataset
     end
 
+    def each(&block)
+      columner_data.each(&block)
+    end
+
+    def [](name)
+      columner_data[name.to_sym]
+    end
+
+    def fetch_values(*keys)
+      data = columner_data
+      keys.collect do |key|
+        if data.key?(key)
+          data[key]
+        else
+          raise build_key_error(key) unless block_given?
+          yield(key)
+        end
+      end
+    end
+
     def to_h
       columns = {}
       @dataset.each do |record|
@@ -14,5 +36,24 @@ module Datasets
       end
       columns
     end
+
+    private
+    begin
+      KeyError.new("message", receiver: self, key: :key)
+    rescue ArgumentError
+      def build_key_error(key)
+        KeyError.new("key not found: #{key.inspect}")
+      end
+    else
+      def build_key_error(key)
+        KeyError.new("key not found: #{key.inspect}",
+                     receiver: self,
+                     key: key)
+      end
+    end
+
+    def columner_data
+      @columns ||= to_h
+    end
   end
 end

data/lib/datasets/version.rb

@@ -1,3 +1,3 @@
 module Datasets
-  VERSION = "0.0.4"
+  VERSION = "0.0.5"
 end

data/test/test-cifar.rb

@@ -168,6 +168,11 @@ class CIFARTest < Test::Unit::TestCase
                      ],
                      raw_dataset)
       end
+
+      test("#to_table") do
+        assert_equal([[10] * 3072, [20] * 3072],
+                     @dataset.to_table[:pixels])
+      end
     end
 
     sub_test_case("test") do
@@ -204,6 +209,11 @@ class CIFARTest < Test::Unit::TestCase
                      ],
                      raw_dataset)
       end
+
+      test("#to_table") do
+        assert_equal([[100] * 3072, [200] * 3072],
+                     @dataset.to_table[:pixels])
+      end
     end
   end
 end

data/test/test-table.rb

@@ -3,6 +3,56 @@ class TableTest < Test::Unit::TestCase
     @table = Datasets::Iris.new.to_table
   end
 
+  test("#[]") do
+    assert_equal([1.4, 1.4, 1.3, 1.5, 1.4],
+                 @table[:petal_length].first(5))
+  end
+
+  sub_test_case("#fetch_values") do
+    test("found") do
+      values = @table.fetch_values(:petal_length, :petal_width)
+      assert_equal([
+                     [1.4, 1.4, 1.3, 1.5, 1.4],
+                     [0.2, 0.2, 0.2, 0.2, 0.2],
+                   ],
+                   values.collect {|v| v.first(5)})
+    end
+
+    sub_test_case("not found") do
+      test("with block") do
+        values = @table.fetch_values(:petal_length, :unknown) do |key|
+          [key] * 5
+        end
+        assert_equal([
+                       [1.4, 1.4, 1.3, 1.5, 1.4],
+                       [:unknown] * 5,
+                     ],
+                     values.collect {|v| v.first(5)})
+      end
+
+      test("without block") do
+        assert_raise(KeyError) do
+          @table.fetch_values(:unknown)
+        end
+      end
+    end
+  end
+
+  test("#each") do
+    shorten_hash = {}
+    @table.each do |name, values|
+      shorten_hash[name] = values.first(5)
+    end
+    assert_equal({
+                   :class        => ["Iris-setosa"] * 5,
+                   :petal_length => [1.4, 1.4, 1.3, 1.5, 1.4],
+                   :petal_width  => [0.2, 0.2, 0.2, 0.2, 0.2],
+                   :sepal_length => [5.1, 4.9, 4.7, 4.6, 5.0],
+                   :sepal_width  => [3.5, 3.0, 3.2, 3.1, 3.6],
+                 },
+                 shorten_hash)
+  end
+
   test("#to_h") do
     shorten_hash = {}
     @table.to_h.each do |name, values|

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: red-datasets
 version: !ruby/object:Gem::Version
-  version: 0.0.4
+  version: 0.0.5
 platform: ruby
 authors:
 - tomisuker
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-05-03 00:00:00.000000000 Z
+date: 2018-06-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -134,7 +134,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.7.6
+rubygems_version: 3.0.0.beta1
 signing_key: 
 specification_version: 4
 summary: Red Datasets provides classes that provide common datasets such as iris dataset.