sooth 1.0.3 → 2.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 85592b0c52f53800465928af00ab97da342b71d4
-  data.tar.gz: dd9c4276380250e4cc0219264e67d138dcc265ee
+  metadata.gz: 9bac8ab9338973373ec7328dd58dcae462e86485
+  data.tar.gz: ea7f2ecb422e0301e36d5f6826aa8ccdbaefb619
 SHA512:
-  metadata.gz: d8969158143b9e4af257c881abccfa0d0624b25e1232ad4fa6f2cde5ba4e5ca154afbbf52fa4f97eca274f6bf3d7d64389d0637e2f05eddb82e32ebcf19bfed0
-  data.tar.gz: 1f9a33941477759b9bbb349d203b1a40d5bdfa18f627fe75e175beba836111be8d28ca405e96ec018b40c43078f2607b6f7687eb60519945f2cb676cb7ad271d
+  metadata.gz: bf39ddf48fba680f4e6f84d42d5f9dc2f59999c83aabafc9f6ba8761f77d24835cfd5b6a6e9f349d61c0e44caec9b7ce2276ba181e4cb13c96980f177481e5ee
+  data.tar.gz: 4ab08bdf68a0f38b676c033f54949bf6a6854f2ade41a2f32b2a62b8a3cff49dcd58b3d89b3f6afa60cef5e6c3863fd22d4a6afd6223285844e72bef4b63d24c

data/CHANGELOG.md

@@ -1,35 +1,30 @@
 # Change Log
 
-## [v1.0.2](https://github.com/jasonhutchens/sooth/tree/v1.0.2) (2015-05-26)
+## [v1.0.3](https://github.com/jasonhutchens/sooth/tree/v1.0.3) (2015-10-03)
+[Full Changelog](https://github.com/jasonhutchens/sooth/compare/v1.0.2...v1.0.3)
 
+## [v1.0.2](https://github.com/jasonhutchens/sooth/tree/v1.0.2) (2015-05-26)
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v1.0.1...v1.0.2)
 
 ## [v1.0.1](https://github.com/jasonhutchens/sooth/tree/v1.0.1) (2015-05-26)
-
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v1.0.0...v1.0.1)
 
 ## [v1.0.0](https://github.com/jasonhutchens/sooth/tree/v1.0.0) (2014-12-17)
-
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v0.5.0...v1.0.0)
 
 ## [v0.5.0](https://github.com/jasonhutchens/sooth/tree/v0.5.0) (2014-12-16)
-
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v0.4.0...v0.5.0)
 
 ## [v0.4.0](https://github.com/jasonhutchens/sooth/tree/v0.4.0) (2014-12-16)
-
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v0.3.0...v0.4.0)
 
 ## [v0.3.0](https://github.com/jasonhutchens/sooth/tree/v0.3.0) (2014-12-11)
-
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v0.2.0...v0.3.0)
 
 ## [v0.2.0](https://github.com/jasonhutchens/sooth/tree/v0.2.0) (2014-12-10)
-
 [Full Changelog](https://github.com/jasonhutchens/sooth/compare/v0.1.0...v0.2.0)
 
 ## [v0.1.0](https://github.com/jasonhutchens/sooth/tree/v0.1.0) (2014-12-10)
 
 
-
 \* *This Change Log was automatically generated by [github_changelog_generator](https://github.com/skywinder/Github-Changelog-Generator)*

data/Gemfile

@@ -7,7 +7,7 @@ group :development do
   gem 'bundler', '~> 1.7'
   gem 'jeweler', '~> 2.0'
   gem 'simplecov', '~> 0.9'
-  gem 'byebug', '~> 6.0'
+  gem 'byebug', '~> 8.2'
   gem 'rake-compiler', '~> 0.9'
   gem 'github_changelog_generator', '~> 1.4'
 end

data/Gemfile.lock

@@ -1,31 +1,40 @@
 GEM
   remote: https://rubygems.org/
   specs:
-    addressable (2.3.8)
+    addressable (2.4.0)
+    ast (2.2.0)
     builder (3.2.2)
-    byebug (6.0.2)
+    byebug (8.2.2)
+    childprocess (0.5.9)
+      ffi (~> 1.0, >= 1.0.11)
     colorize (0.7.7)
     descendants_tracker (0.0.4)
       thread_safe (~> 0.3, >= 0.3.1)
     diff-lcs (1.2.5)
     docile (1.1.5)
-    faraday (0.9.1)
+    faraday (0.9.2)
       multipart-post (>= 1.2, < 3)
-    git (1.2.9.1)
-    github_api (0.12.4)
-      addressable (~> 2.3)
+    ffi (1.9.10)
+    git (1.3.0)
+    github_api (0.13.1)
+      addressable (~> 2.4.0)
       descendants_tracker (~> 0.0.4)
       faraday (~> 0.8, < 0.10)
       hashie (>= 3.4)
       multi_json (>= 1.7.5, < 2.0)
-      nokogiri (~> 1.6.6)
       oauth2
-    github_changelog_generator (1.9.0)
+    github_changelog_generator (1.12.0)
+      bundler (>= 1.7)
       colorize (~> 0.7)
       github_api (~> 0.12)
-    hashie (3.4.2)
-    highline (1.7.7)
-    jeweler (2.0.1)
+      overcommit (>= 0.31)
+      rake (>= 10.0)
+      rspec (>= 3.2)
+      rubocop (>= 0.31)
+    hashie (3.4.3)
+    highline (1.7.8)
+    iniparse (1.4.2)
+    jeweler (2.1.1)
       builder
       bundler (>= 1.0)
       git (>= 1.2.5)
@@ -34,44 +43,62 @@ GEM
       nokogiri (>= 1.5.10)
       rake
       rdoc
+      semver
     json (1.8.3)
     jwt (1.5.1)
-    mini_portile (0.6.2)
+    mini_portile2 (2.0.0)
     multi_json (1.11.2)
     multi_xml (0.5.5)
     multipart-post (2.0.0)
-    nokogiri (1.6.6.2)
-      mini_portile (~> 0.6.0)
-    oauth2 (1.0.0)
+    nokogiri (1.6.7.2)
+      mini_portile2 (~> 2.0.0.rc2)
+    oauth2 (1.1.0)
       faraday (>= 0.8, < 0.10)
-      jwt (~> 1.0)
+      jwt (~> 1.0, < 1.5.2)
       multi_json (~> 1.3)
       multi_xml (~> 0.5)
-      rack (~> 1.2)
+      rack (>= 1.2, < 3)
+    overcommit (0.33.0)
+      childprocess (~> 0.5.8)
+      iniparse (~> 1.4)
+    parser (2.3.0.7)
+      ast (~> 2.2)
+    powerpack (0.1.1)
     rack (1.6.4)
-    rake (10.4.2)
-    rake-compiler (0.9.5)
+    rainbow (2.1.0)
+    rake (11.1.2)
+    rake-compiler (0.9.7)
       rake
-    rdoc (4.2.0)
-    rspec (3.3.0)
-      rspec-core (~> 3.3.0)
-      rspec-expectations (~> 3.3.0)
-      rspec-mocks (~> 3.3.0)
-    rspec-core (3.3.2)
-      rspec-support (~> 3.3.0)
-    rspec-expectations (3.3.1)
+    rdoc (4.2.2)
+      json (~> 1.4)
+    rspec (3.4.0)
+      rspec-core (~> 3.4.0)
+      rspec-expectations (~> 3.4.0)
+      rspec-mocks (~> 3.4.0)
+    rspec-core (3.4.4)
+      rspec-support (~> 3.4.0)
+    rspec-expectations (3.4.0)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.3.0)
-    rspec-mocks (3.3.2)
+      rspec-support (~> 3.4.0)
+    rspec-mocks (3.4.1)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.3.0)
-    rspec-support (3.3.0)
-    simplecov (0.10.0)
+      rspec-support (~> 3.4.0)
+    rspec-support (3.4.1)
+    rubocop (0.39.0)
+      parser (>= 2.3.0.7, < 3.0)
+      powerpack (~> 0.1)
+      rainbow (>= 1.99.1, < 3.0)
+      ruby-progressbar (~> 1.7)
+      unicode-display_width (~> 1.0, >= 1.0.1)
+    ruby-progressbar (1.7.5)
+    semver (1.0.1)
+    simplecov (0.11.2)
       docile (~> 1.1.0)
       json (~> 1.8)
       simplecov-html (~> 0.10.0)
     simplecov-html (0.10.0)
     thread_safe (0.3.5)
+    unicode-display_width (1.0.3)
     yard (0.8.7.6)
 
 PLATFORMS
@@ -79,7 +106,7 @@ PLATFORMS
 
 DEPENDENCIES
   bundler (~> 1.7)
-  byebug (~> 6.0)
+  byebug (~> 8.2)
   github_changelog_generator (~> 1.4)
   jeweler (~> 2.0)
   rake-compiler (~> 0.9)
@@ -89,4 +116,4 @@ DEPENDENCIES
   yard (~> 0.8)
 
 BUNDLED WITH
-   1.10.5
+   1.11.2

data/README.md

@@ -5,15 +5,18 @@
 Sooth
 =====
 
-Sooth is a simple stochastic predictive model. It is used by
-[MegaHAL](https://github.com/jasonhutchens/megahal), a learning chatterbot.
+Sooth is a minimal stochastic predictive model. It is used by [MegaHAL](https://github.com/jasonhutchens/megahal), a learning chatterbot, and by [Typing Simulator](https://github.com/jasonhutchens/typing_simulator), a program that pretends to type text files like a human being.
 
 Getting Started
 ---------------
 
-Apart from the specs, look at the [API docs](http://www.rubydoc.info/gems/sooth/) to get up to speed.
+The basic philisophy is to keep things simple. Sooth can make _predictions_ about _events_ that occur within some _context_.
+
+A _context_ is just a number; it's up to you to make it significant. It might represent a word, or an array of words, or something else entirely. Sooth doesn't care. Likewise, an _event_ is a number, but doesn't need to represent the same kind of thing as a _context_. It's perfectly fine for a _context_ to be an array of words and an _event_ to be a single word, which is what MegaHAL does. Or a _context_ could be an array of characters and an _event_ could be a time in milliseconds, which is what TypingSimulator does. Or they could be something else entirely...
+
+Look at the [API docs](http://www.rubydoc.info/gems/sooth/Sooth/Predictor) to get up to speed, and at the specs for example usage.
 
 Copyright
 ---------
 
-Copyright (c) 2015 Jason Hutchens. See UNLICENSE for further details.
+Copyright (c) 2016 Jason Hutchens. See UNLICENSE for further details.

data/VERSION

@@ -1 +1 @@
-1.0.3
+2.0.0

data/ext/sooth_native/extconf.rb

@@ -1,5 +1,5 @@
 require 'mkmf'
 
-$CFLAGS << ' -Wall -O3 -std=gnu99'
+$CFLAGS << ' -Wall -Ofast -std=gnu99'
 
 create_makefile('sooth_native')

data/ext/sooth_native/native.c

@@ -14,11 +14,10 @@ void method_sooth_native_deallocate(void * predictor);
 
 /* @!parse [ruby]
  *   module Sooth
- *     # A very simple stochastic predictor. Implemented in C for efficiency.
- *     # The idea here is to build up more complicated learning algorithms using
- *     # a trivial Markovian predictor.
+ *     # A minimal stochastic predictive model, implemented in C for efficiency.
+ *     # No assumptions about PRNG or real-world significance of context/event.
  *     class Predictor
- *       def initialize(error_symbol)
+ *       def initialize(error_event)
  *       end
  *       def clear
  *       end
@@ -26,19 +25,25 @@ void method_sooth_native_deallocate(void * predictor);
  *       end
  *       def save(filename)
  *       end
- *       def observe(bigram, symbol)
+ *       def size(context)
  *         # (native code)
  *       end
- *       def count(bigram)
+ *       def count(context)
  *         # (native code)
  *       end
- *       def select(bigram, limit)
+ *       def observe(context, event)
  *         # (native code)
  *       end
- *       def uncertainty(bigram)
+ *       def select(context, limit)
  *         # (native code)
  *       end
- *       def surprise(bigram, symbol)
+ *       def distribution(context)
+ *         # (native code)
+ *       end
+ *       def uncertainty(context)
+ *         # (native code)
+ *       end
+ *       def surprise(context, evemt)
  *         # (native code)
  *       end
  *     end
@@ -46,13 +51,13 @@ void method_sooth_native_deallocate(void * predictor);
  *
  * Returns a new Sooth::Predictor instance.
  *
- * @param [Fixnum] error_symbol The symbol to be returned by #select when no
- *                              prediction can be made.
+ * @param [Fixnum] error_event The event to be returned by #select when no
+ *                             observations have been made for the context.
  */
-VALUE method_sooth_native_initialize(VALUE self, VALUE error_symbol);
+VALUE method_sooth_native_initialize(VALUE self, VALUE error_event);
 
 /*
- * Clear the predictor to a fresh slate.
+ * Clear the predictor to a blank slate.
  */
 VALUE method_sooth_native_clear(VALUE self);
 
@@ -65,76 +70,92 @@ VALUE method_sooth_native_clear(VALUE self);
 VALUE method_sooth_native_load(VALUE self, VALUE filename);
 
 /*
- * Save the predictor to a file that can be loaded or merged later.
+ * Save the predictor to a file that can be loaded later.
  *
  * @param [String] filename The path of the file to be merge.
  */
 VALUE method_sooth_native_save(VALUE self, VALUE filename);
 
 /*
- * Add an observation of the given symbol in the context of the bigram.
+ * Return the number of different events that have been observed within the
+ * given context.
  *
- * @param [Array] bigram A pair of symbols that provide context, allowing the
- *                       predictor to maintain observation statistics for
- *                       different contexts.
- * @param [Fixnum] symbol The symbol that has been observed.
- * @return [Fixnum] A count of the number of times the symbol has been
- *                  observed in the context of the bigram.
+ * @param [Fixnum] context A number that provides a context for observations.
+ * @return [Fixnum] The number of distinct events that have been observed
+ *                  within the given context. This is guaranteed to be equal
+ *                  to the length of the #distribution for the context.
  */
-VALUE method_sooth_native_observe(VALUE self, VALUE bigram, VALUE symbol);
+
+VALUE method_sooth_native_size(VALUE self, VALUE context);
 
 /*
- * Return a count of the number of times the bigram has been observed.
+ * Return the number of times the context has been observed.
  *
- * @param [Array] bigram A pair of symbols.
- * @return [Fixnum] A count of the number of times the bigram has been
+ * @param [Fixnum] context A number that provides a context for observations.
+ * @return [Fixnum] A count of the number of times the context has been
  *                  observed. This is guaranteed to be equal to the sum
- *                  of the counts of observations of all symbols in the
- *                  context of the bigram.
+ *                  of the counts of observations of all events observed in
+ *                  the context.
+ */
+VALUE method_sooth_native_count(VALUE self, VALUE context);
+
+/*
+ * Register an observation of the given event within the given context.
+ *
+ * @param [Fixnum] context A number that provides a context for the event,
+ *                         allowing the predictor to maintain observation
+ *                         statistics for different contexts.
+ * @param [Fixnum] event A number representing the observed event.
+ * @return [Fixnum] A count of the number of times the event has been
+ *                  observed in the given context.
+ */
+VALUE method_sooth_native_observe(VALUE self, VALUE context, VALUE event);
+
+/*
+ * Return an event that may occur in the given context, based on the limit,
+ * which should be between 1 and #count. The event is selected by iterating
+ * through all observed events for the context, subtracting the observation
+ * count of each event from the limit until it is zero or less.
+ *
+ * @param [Fixnum] context A number that provides a context for observations.
+ * @param [Fixnum] limit The total number of event observations to be
+ *                       analysed before returning a event.
+ * @return [Fixnum] An event that has been previously observed in the given
+ *                  context, or the error_event if the #count of the context
+ *                  is zero, or if limit exceeds the #count of the context
  */
-VALUE method_sooth_native_count(VALUE self, VALUE bigram);
+VALUE method_sooth_native_select(VALUE self, VALUE context, VALUE limit);
 
 /*
- * Return a symbol that may occur in the context of the bigram. The
- * limit is used to select a symbol. This is done by iterating through
- * all of the symbols that have been observed in the context of the
- * bigram, subtracting the observation count of each symbol from the
- * supplied limit. For this reason, limit should be between 1 and the
- * observation count of the bigram itself, as returned by #count.
+ * Return an Enumerator that yields each observed event within the context
+ * together with its probability.
  *
- * @param [Array] bigram A pair of symbols.
- * @param [Fixnum] limit The total numbe of symbol observations to be
- *                       analysed before returning a symbol.
- * @return [Fixnum] A symbol that has been observed previously in the
- *                  context of the bigram, or the error_symbol if no
- *                  such symbol exists, or if the supplied limit was
- *                  too large.
+ * @param [Fixnum] context A number that provides a context for observations.
+ * @return [Array] A list of event-probability pairs.
  */
-VALUE method_sooth_native_select(VALUE self, VALUE bigram, VALUE limit);
+VALUE method_sooth_native_distribution(VALUE self, VALUE context);
 
 /*
- * Return a number indicating how uncertain the predictor is about which symbol
- * is likely to be observed after the given bigram. Note that nil will be
- * returned if the bigram has never been observed.
+ * Return a number indicating how uncertain the predictor is about which event
+ * is likely to be observed after the given context. Note that nil will be
+ * returned if the context has never been observed.
  *
- * @param [Array] bigram A pair of symbols.
- * @return [Float] The uncertainty, which is calculated to be the shannon entropy
- *                 of the probability distribution over the alphabet of symbols
- *                 in the context of the bigram.
+ * @param [Fixnum] context A number that provides a context for observations.
+ * @return [Float] The uncertainty, which is calculated to be the Shannon entropy
+ *                 of the #distribution over the context.
  */
-VALUE method_sooth_native_uncertainty(VALUE self, VALUE bigram);
+VALUE method_sooth_native_uncertainty(VALUE self, VALUE context);
 
 /*
  * Return a number indicating the surprise received by the predictor when it
- * observed the given symbol after the given bigram. Note that nil will be
- * returned if the symbol has never been observed after the bigram.
+ * observed the given event within the given context. Note that nil will be
+ * returned if the event has never been observed within the context.
  *
- * @param [Array] bigram A pair of symbols.
- * @param [Fixnum] symbol The symbol that has been observed.
- * @return [Float] The surprise, which is calculated to be the shannon pointwise
- *                 mutual information of the symbol according to the probability
- *                 distribution over the alphabet of symbols in the context of
- *                 the bigram.
+ * @param [Fixnum] context A number that provides a context for observations.
+ * @param [Fixnum] event A number representing the observed event.
+ * @return [Float] The surprise, which is calculated to be the Shannon pointwise
+ *                 mutual information of the event according to the
+ *                 #distribution over the context.
  */
 VALUE method_sooth_native_surprise(VALUE self, VALUE bigram, VALUE limit);
 
@@ -152,9 +173,12 @@ void Init_sooth_native()
   rb_define_method(SoothNative, "load", method_sooth_native_load, 1);
   rb_define_method(SoothNative, "save", method_sooth_native_save, 1);
 
-  rb_define_method(SoothNative, "observe", method_sooth_native_observe, 2);
+  rb_define_method(SoothNative, "size", method_sooth_native_size, 1);
   rb_define_method(SoothNative, "count", method_sooth_native_count, 1);
+  rb_define_method(SoothNative, "observe", method_sooth_native_observe, 2);
   rb_define_method(SoothNative, "select", method_sooth_native_select, 2);
+
+  rb_define_method(SoothNative, "distribution", method_sooth_native_distribution, 1);
   rb_define_method(SoothNative, "uncertainty", method_sooth_native_uncertainty, 1);
   rb_define_method(SoothNative, "surprise", method_sooth_native_surprise, 2);
 }
@@ -183,12 +207,12 @@ method_sooth_native_deallocate(void * predictor)
 //------------------------------------------------------------------------------
 
 VALUE
-method_sooth_native_initialize(VALUE self, VALUE error_symbol)
+method_sooth_native_initialize(VALUE self, VALUE error_event)
 {
   SoothPredictor * predictor = NULL;
-  Check_Type(error_symbol, T_FIXNUM);
+  Check_Type(error_event, T_FIXNUM);
   Data_Get_Struct(self, SoothPredictor, predictor);
-  predictor->error_symbol = NUM2UINT(error_symbol);
+  predictor->error_event = NUM2UINT(error_event);
   return self;
 }
 
@@ -236,78 +260,90 @@ method_sooth_native_save(VALUE self, VALUE filename)
 //------------------------------------------------------------------------------
 
 VALUE
-method_sooth_native_observe(VALUE self, VALUE bigram, VALUE symbol)
+method_sooth_native_size(VALUE self, VALUE context)
 {
   SoothPredictor * predictor = NULL;
-  Check_Type(symbol, T_FIXNUM);
-  Check_Type(bigram, T_ARRAY);
-  if (RARRAY_LEN(bigram) != 2)
-  {
-    rb_raise(rb_eTypeError, "bigram must be an array of exactly two symbols");
-  }
-  Check_Type(RARRAY_PTR(bigram)[0], T_FIXNUM);
-  Check_Type(RARRAY_PTR(bigram)[1], T_FIXNUM);
+  Check_Type(context, T_FIXNUM);
   Data_Get_Struct(self, SoothPredictor, predictor);
-  uint32_t c_bigram[2] = {NUM2UINT(RARRAY_PTR(bigram)[0]), NUM2UINT(RARRAY_PTR(bigram)[1])};
-  uint32_t count = sooth_predictor_observe(predictor, c_bigram, NUM2UINT(symbol));
+  uint32_t size = sooth_predictor_size(predictor, NUM2UINT(context));
+  return UINT2NUM(size);
+}
+
+//------------------------------------------------------------------------------
+
+VALUE
+method_sooth_native_count(VALUE self, VALUE context)
+{
+  SoothPredictor * predictor = NULL;
+  Check_Type(context, T_FIXNUM);
+  Data_Get_Struct(self, SoothPredictor, predictor);
+  uint32_t count = sooth_predictor_count(predictor, NUM2UINT(context));
   return UINT2NUM(count);
 }
 
 //------------------------------------------------------------------------------
 
 VALUE
-method_sooth_native_count(VALUE self, VALUE bigram)
+method_sooth_native_observe(VALUE self, VALUE context, VALUE event)
 {
   SoothPredictor * predictor = NULL;
-  Check_Type(bigram, T_ARRAY);
-  if (RARRAY_LEN(bigram) != 2)
-  {
-    rb_raise(rb_eTypeError, "bigram must be an array of exactly two symbols");
-  }
-  Check_Type(RARRAY_PTR(bigram)[0], T_FIXNUM);
-  Check_Type(RARRAY_PTR(bigram)[1], T_FIXNUM);
+  Check_Type(context, T_FIXNUM);
+  Check_Type(event, T_FIXNUM);
   Data_Get_Struct(self, SoothPredictor, predictor);
-  uint32_t c_bigram[2] = {NUM2UINT(RARRAY_PTR(bigram)[0]), NUM2UINT(RARRAY_PTR(bigram)[1])};
-  uint32_t count = sooth_predictor_count(predictor, c_bigram);
+  uint32_t count = sooth_predictor_observe(predictor, NUM2UINT(context), NUM2UINT(event));
   return UINT2NUM(count);
 }
 
 //------------------------------------------------------------------------------
 
 VALUE
-method_sooth_native_select(VALUE self, VALUE bigram, VALUE limit)
+method_sooth_native_select(VALUE self, VALUE context, VALUE limit)
 {
   SoothPredictor * predictor = NULL;
+  Check_Type(context, T_FIXNUM);
   Check_Type(limit, T_FIXNUM);
-  Check_Type(bigram, T_ARRAY);
-  if (RARRAY_LEN(bigram) != 2)
-  {
-    rb_raise(rb_eTypeError, "bigram must be an array of exactly two symbols");
-  }
-  Check_Type(RARRAY_PTR(bigram)[0], T_FIXNUM);
-  Check_Type(RARRAY_PTR(bigram)[1], T_FIXNUM);
   Data_Get_Struct(self, SoothPredictor, predictor);
-  uint32_t c_bigram[2] = {NUM2UINT(RARRAY_PTR(bigram)[0]), NUM2UINT(RARRAY_PTR(bigram)[1])};
-  uint32_t symbol = sooth_predictor_select(predictor, c_bigram, NUM2UINT(limit));
-  return UINT2NUM(symbol);
+  uint32_t event = sooth_predictor_select(predictor, NUM2UINT(context), NUM2UINT(limit));
+  return UINT2NUM(event);
 }
 
 //------------------------------------------------------------------------------
 
 VALUE
-method_sooth_native_uncertainty(VALUE self, VALUE bigram)
+method_sooth_native_distribution(VALUE self, VALUE context)
 {
   SoothPredictor * predictor = NULL;
-  Check_Type(bigram, T_ARRAY);
-  if (RARRAY_LEN(bigram) != 2)
+  Check_Type(context, T_FIXNUM);
+  Data_Get_Struct(self, SoothPredictor, predictor);
+  uint32_t c_context = NUM2UINT(context);
+  SoothStatistic * statistics = sooth_predictor_distribution(predictor, c_context);
+  if (statistics == NULL)
   {
-    rb_raise(rb_eTypeError, "bigram must be an array of exactly two symbols");
+    return Qnil;
   }
-  Check_Type(RARRAY_PTR(bigram)[0], T_FIXNUM);
-  Check_Type(RARRAY_PTR(bigram)[1], T_FIXNUM);
+  uint32_t size = sooth_predictor_size(predictor, c_context);
+  double count = (double)sooth_predictor_count(predictor, c_context);
+  VALUE r_array = rb_ary_new2(size);
+  for (uint32_t i = 0; i < size; ++i)
+  {
+    SoothStatistic statistic = statistics[i];
+    VALUE pair = rb_ary_new2(2);
+    rb_ary_store(pair, 0, UINT2NUM(statistic.event));
+    rb_ary_store(pair, 1, DBL2NUM((double)statistic.count/count));
+    rb_ary_store(r_array, i, pair);
+  }
+  return r_array;
+}
+
+//------------------------------------------------------------------------------
+
+VALUE
+method_sooth_native_uncertainty(VALUE self, VALUE context)
+{
+  SoothPredictor * predictor = NULL;
+  Check_Type(context, T_FIXNUM);
   Data_Get_Struct(self, SoothPredictor, predictor);
-  uint32_t c_bigram[2] = {NUM2UINT(RARRAY_PTR(bigram)[0]), NUM2UINT(RARRAY_PTR(bigram)[1])};
-  double uncertainty = sooth_predictor_uncertainty(predictor, c_bigram);
+  double uncertainty = sooth_predictor_uncertainty(predictor, NUM2UINT(context));
   if (uncertainty < 0)
   {
     return Qnil;
@@ -318,20 +354,13 @@ method_sooth_native_uncertainty(VALUE self, VALUE bigram)
 //------------------------------------------------------------------------------
 
 VALUE
-method_sooth_native_surprise(VALUE self, VALUE bigram, VALUE symbol)
+method_sooth_native_surprise(VALUE self, VALUE context, VALUE event)
 {
   SoothPredictor * predictor = NULL;
-  Check_Type(symbol, T_FIXNUM);
-  Check_Type(bigram, T_ARRAY);
-  if (RARRAY_LEN(bigram) != 2)
-  {
-    rb_raise(rb_eTypeError, "bigram must be an array of exactly two symbols");
-  }
-  Check_Type(RARRAY_PTR(bigram)[0], T_FIXNUM);
-  Check_Type(RARRAY_PTR(bigram)[1], T_FIXNUM);
+  Check_Type(context, T_FIXNUM);
+  Check_Type(event, T_FIXNUM);
   Data_Get_Struct(self, SoothPredictor, predictor);
-  uint32_t c_bigram[2] = {NUM2UINT(RARRAY_PTR(bigram)[0]), NUM2UINT(RARRAY_PTR(bigram)[1])};
-  double surprise = sooth_predictor_surprise(predictor, c_bigram, NUM2UINT(symbol));
+  double surprise = sooth_predictor_surprise(predictor, NUM2UINT(context), NUM2UINT(event));
   if (surprise < 0)
   {
     return Qnil;