word2vec-rb 0.4.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f1cc8aad12e17e08da6428945fa3fcdab514739a5ff166588796b0de4535ba10
-  data.tar.gz: 3deaf3e6d6f1bff3e06b6353d8c29591fa00c8f5e8c8f3079f0d9fdf029786b7
+  metadata.gz: 192304d569b2fb573300a33ff7ed05fc37202c30cf55144fc043cc544e620f23
+  data.tar.gz: e5473521b82560242b19ae2efa9c4e07a7798bb3c0ea2e97d97e5f3a1f2601cf
 SHA512:
-  metadata.gz: 916361e4543fadac744e4502f53241b105789bbf928be51d2e17df93b3ac318e2b9125d7146ec67146a591ca87ee7d2928ca15aadb1fabc0b2c31b82e058497b
-  data.tar.gz: 9daa3535f9c23254b5fa45e9759fab036a297ffa491e3fc03287c3c9c072c7b314925843d376e678f1b8f61915839f5e0689e1fa27268704eaf5bd086fdaecc3
+  metadata.gz: db2cc2396b4a7193a0cb4efd40ffefbefabd88861f257aea8ac90d032d9ebac8f0759cd952678174f764cdce2605a94c521e24d9433a8a1eaaf135e5dcc1cf89
+  data.tar.gz: 26949750236e3929f9794967570c3fb257c53474c9d79b857f12d4214ecca075afe4a32dbb5a6e5253b44ea1566a391aff70c9617a2914ee700779e3dd8a4af9

data/.gitignore

@@ -13,4 +13,5 @@
 lib/word2vec/word2vec.bundle
 
 data/text7
-data/vocab.txt
+data/vocab.txt
+data/tokenized.bin

data/.vscode/c_cpp_properties.json

@@ -0,0 +1,20 @@
+{
+    "configurations": [
+        {
+            "name": "Mac",
+            "includePath": [
+                "${workspaceFolder}/**",
+                "/Library/Developer/CommandLineTools/SDKs/MacOSX11.3.sdk/System/Library/Frameworks/Ruby.framework/Versions/2.6/usr/include/ruby-2.6.0/**"
+            ],
+            "defines": [],
+            "macFrameworkPath": [
+                "/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks"
+            ],
+            "compilerPath": "/usr/bin/clang",
+            "cStandard": "c17",
+            "cppStandard": "c++98",
+            "intelliSenseMode": "macos-clang-x64"
+        }
+    ],
+    "version": 4
+}

data/.vscode/settings.json

@@ -0,0 +1,6 @@
+{
+    "files.associations": {
+        "common.h": "c",
+        "ruby.h": "c"
+    }
+}

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    word2vec-rb (0.4.0)
+    word2vec-rb (0.5.0)
 
 GEM
   remote: https://rubygems.org/

data/README.md

@@ -78,6 +78,23 @@ Word2vec::Model.build\_vocab("./data/text7", "./data/vocab.txt")
 
 The output file will have a list of words and its number of appearances separated by line break.
 
+### Tokenizer: create a binary file by tokenizing an input file
+
+This method requires a vocabulary file precreated.
+
+```ruby
+require 'word2vec'
+
+Word2vec::Model.tokenize("./data/text7", "./data/vocab.txt", "./data/toekized.bin")
+```
+
+The output file will contain a sequence of binary identificators of each word of the input file.
+
+Read output file with:
+
+    long long id;
+    fread(&id, sizeof(id), 1, fi);
+
 ## Development
 
 After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.

data/ext/word2vec/common.c

@@ -1,5 +1,7 @@
 #include "common.h"
 
+char train_file[MAX_STRING];
+char read_vocab_file[MAX_STRING];
 // max length of strings
 const unsigned long max_size = 2000;
 // number of closest words that will be shown
@@ -44,4 +46,199 @@ void word2vec_model_load(word2vec_model* model, char* file_name) {
     for (a = 0; a < size; a++) model->vectors[a + b * size] /= len;
   }
   fclose(f);
-}
+}
+
+// Reads a single word from a file, assuming space + tab + EOL to be word boundaries
+void ReadWord(char *word, FILE *fin) {
+  int a = 0, ch;
+  while (!feof(fin)) {
+    ch = fgetc(fin);
+    if (ch == 13) continue;
+    if ((ch == ' ') || (ch == '\t') || (ch == '\n')) {
+      if (a > 0) {
+        if (ch == '\n') ungetc(ch, fin);
+        break;
+      }
+      if (ch == '\n') {
+        strcpy(word, (char *)"</s>");
+        return;
+      } else continue;
+    }
+    word[a] = ch;
+    a++;
+    if (a >= MAX_STRING - 1) a--;   // Truncate too long words
+  }
+  word[a] = 0;
+}
+
+// Returns hash value of a word
+unsigned long long GetWordHash(char *word) {
+  unsigned long long a, hash = 0;
+  for (a = 0; a < strlen(word); a++) hash = hash * 257 + word[a];
+  hash = hash % vocab_hash_size;
+  return hash;
+}
+
+// Returns position of a word in the vocabulary; if the word is not found, returns -1
+unsigned long long SearchVocab(char *word) {
+  unsigned long long hash = GetWordHash(word);
+  while (1) {
+    if (vocab_hash[hash] == -1) return -1;
+    if (!strcmp(word, vocab[vocab_hash[hash]].word)) return vocab_hash[hash];
+    hash = (hash + 1) % vocab_hash_size;
+  }
+  return -1;
+}
+
+// Reads a word and returns its index in the vocabulary
+unsigned long long ReadWordIndex(FILE *fin) {
+  char word[MAX_STRING];
+  ReadWord(word, fin);
+  if (feof(fin)) return -1;
+  return SearchVocab(word);
+}
+
+// Adds a word to the vocabulary
+unsigned long long AddWordToVocab(char *word) {
+  unsigned long long hash, length = strlen(word) + 1;
+  if (length > MAX_STRING) length = MAX_STRING;
+  vocab[vocab_size].word = (char *)calloc(length, sizeof(char));
+  strcpy(vocab[vocab_size].word, word);
+  vocab[vocab_size].cn = 0;
+  vocab_size++;
+  // Reallocate memory if needed
+  if (vocab_size + 2 >= vocab_max_size) {
+    vocab_max_size += 1000;
+    vocab = (struct vocab_word *)realloc(vocab, vocab_max_size * sizeof(struct vocab_word));
+  }
+  hash = GetWordHash(word);
+  while (vocab_hash[hash] != -1) hash = (hash + 1) % vocab_hash_size;
+  vocab_hash[hash] = vocab_size - 1;
+  return vocab_size - 1;
+}
+
+// Reduces the vocabulary by removing infrequent tokens
+void ReduceVocab() {
+  int a, b = 0;
+  unsigned long long hash;
+  for (a = 0; a < vocab_size; a++) if (vocab[a].cn > min_reduce) {
+    vocab[b].cn = vocab[a].cn;
+    vocab[b].word = vocab[a].word;
+    b++;
+  } else free(vocab[a].word);
+  vocab_size = b;
+  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
+  for (a = 0; a < vocab_size; a++) {
+    // Hash will be re-computed, as it is not actual
+    hash = GetWordHash(vocab[a].word);
+    while (vocab_hash[hash] != -1) hash = (hash + 1) % vocab_hash_size;
+    vocab_hash[hash] = a;
+  }
+  fflush(stdout);
+  min_reduce++;
+}
+
+// Used later for sorting by word counts
+int VocabCompare(const void *a, const void *b) {
+    return (int)(((struct vocab_word *)b)->cn - ((struct vocab_word *)a)->cn);
+}
+
+// Sorts the vocabulary by frequency using word counts
+void SortVocab() {
+  long long a, size;
+  unsigned long long hash;
+  // Sort the vocabulary and keep </s> at the first position
+  qsort(&vocab[1], vocab_size - 1, sizeof(struct vocab_word), VocabCompare);
+  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
+  size = vocab_size;
+  train_words = 0;
+  for (a = 0; a < size; a++) {
+    // Words occuring less than min_count times will be discarded from the vocab
+    if ((vocab[a].cn < min_count) && (a != 0)) {
+      vocab_size--;
+      free(vocab[a].word);
+    } else {
+      // Hash will be re-computed, as after the sorting it is not actual
+      hash=GetWordHash(vocab[a].word);
+      while (vocab_hash[hash] != -1) hash = (hash + 1) % vocab_hash_size;
+      vocab_hash[hash] = a;
+      train_words += vocab[a].cn;
+    }
+  }
+  vocab = (struct vocab_word *)realloc(vocab, (vocab_size + 1) * sizeof(struct vocab_word));
+  // Allocate memory for the binary tree construction
+  for (a = 0; a < vocab_size; a++) {
+    vocab[a].code = (char *)calloc(MAX_CODE_LENGTH, sizeof(char));
+    vocab[a].point = (int *)calloc(MAX_CODE_LENGTH, sizeof(int));
+  }
+}
+
+// Learn vocabulary data from file
+void LearnVocabFromTrainFile() {
+  char word[MAX_STRING];
+  FILE *fin;
+  long long a, i;
+  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
+  fin = fopen(train_file, "rb");
+  if (fin == NULL) {
+    printf("ERROR: training data file not found!\n");
+    exit(1);
+  }
+  vocab_size = 0;
+  AddWordToVocab((char *)"</s>");
+  while (1) {
+    ReadWord(word, fin);
+    if (feof(fin)) break;
+    train_words++;
+    if ((debug_mode > 1) && (train_words % 100000 == 0)) {
+      printf("%lldK%c", train_words / 1000, 13);
+      fflush(stdout);
+    }
+    i = SearchVocab(word);
+    if (i == -1) {
+      a = AddWordToVocab(word);
+      vocab[a].cn = 1;
+    } else vocab[i].cn++;
+    if (vocab_size > vocab_hash_size * 0.7) ReduceVocab();
+  }
+  SortVocab();
+  if (debug_mode > 0) {
+    printf("Vocab size: %lld\n", vocab_size);
+    printf("Words in train file: %lld\n", train_words);
+  }
+  file_size = ftell(fin);
+  fclose(fin);
+}
+
+void ReadVocab() {
+  long long a, i = 0;
+  char c;
+  char word[MAX_STRING];
+  FILE *fin = fopen(read_vocab_file, "rb");
+  if (fin == NULL) {
+    printf("Vocabulary file not found\n");
+    exit(1);
+  }
+  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
+  vocab_size = 0;
+  while (1) {
+    ReadWord(word, fin);
+    if (feof(fin)) break;
+    a = AddWordToVocab(word);
+    fscanf(fin, "%lld%c", &vocab[a].cn, &c);
+    i++;
+  }
+  SortVocab();
+  if (debug_mode > 0) {
+    printf("Vocab size: %lld\n", vocab_size);
+    printf("Words in train file: %lld\n", train_words);
+  }
+  fin = fopen(train_file, "rb");
+  if (fin == NULL) {
+    printf("ERROR: training data file not found!\n");
+    exit(1);
+  }
+  fseek(fin, 0, SEEK_END);
+  file_size = ftell(fin);
+  fclose(fin);
+}

data/ext/word2vec/common.h

@@ -10,13 +10,26 @@
 #include <sys/types.h>
 #include <ctype.h>
 
+#define MAX_STRING 100
+#define MAX_CODE_LENGTH 40
+
 // max length of strings
 extern const unsigned long max_size;
 // number of closest words that will be shown
 extern const long long N;
 // max length of vocabulary entries
 extern const long long max_w;
-  
+// Maximum 30 * 0.7 = 21M words in the vocabulary
+extern const int vocab_hash_size;
+extern long long *vocab_hash;
+extern struct vocab_word *vocab;
+
+extern char train_file[];
+extern char read_vocab_file[];
+extern long long vocab_max_size, vocab_size;
+extern long long train_words, file_size ;
+extern int debug_mode, min_count, min_reduce;
+
 typedef struct word2vec_model_s {
   long long word_count;
   char *vocabulary;  // char *[word_count]
@@ -29,10 +42,26 @@ typedef struct WordSimilarity_s {
   float score;
 } WordSimilarity;
 
+struct vocab_word {
+  long long cn;
+  int *point;
+  char *word, *code, codelen;
+};
+
+void ReadWord(char *word, FILE *fin);
+unsigned long long GetWordHash(char *word);
+unsigned long long SearchVocab(char *word) ;
+unsigned long long ReadWordIndex(FILE *fin);
+
+void LearnVocabFromTrainFile();
+void ReadVocab();
+
+/// main functions
 void word2vec_model_load(word2vec_model* model, char* file_name);
 size_t word2vec_model_distance(word2vec_model* model, char* word, WordSimilarity word_list[]);
 size_t word2vec_model_analogy(word2vec_model* model, char* wordx1, char* wordy1, char* wordx2, WordSimilarity word_list[]);
 void word2vec_model_accuracy(word2vec_model* model, char* file_name);
-void word2vec_build_vocab(char* train, char* save_vocab_file);
+void word2vec_build_vocab(char* train_file, char* save_vocab_file);
+void word2vec_tokenize(char* train_file, char* vocab_file, char* output_file);
 
 #endif /* _WORD2VEC_COMMON_H */

data/ext/word2vec/tokenizer.c

@@ -0,0 +1,19 @@
+#include "common.h"
+
+void word2vec_tokenize(char* train_file, char* vocab_file, char* output_file) {
+  long long word;
+  vocab = (struct vocab_word *)calloc(vocab_max_size, sizeof(struct vocab_word));
+  vocab_hash = (long long *)calloc(vocab_hash_size, sizeof(long long));
+  strcpy(train_file, train_file);
+  strcpy(read_vocab_file, vocab_file);
+  ReadVocab();
+  FILE *fi = fopen(train_file, "rb");
+  FILE *fo = fopen(output_file, "wb");
+  while (1) {
+    word = ReadWordIndex(fi);
+    if (feof(fi)) break;
+    fwrite(&word, sizeof(word), 1, fo);
+  }
+  fclose(fi);
+  fclose(fo);
+}

data/ext/word2vec/vocab.c

@@ -1,185 +1,13 @@
 #include "common.h"
 
-#define MAX_STRING 100
-#define MAX_CODE_LENGTH 40
-
 const int vocab_hash_size = 30000000;  // Maximum 30 * 0.7 = 21M words in the vocabulary
 
-struct vocab_word {
-  long long cn;
-  int *point;
-  char *word, *code, codelen;
-};
-
-char train_file[MAX_STRING];
 struct vocab_word *vocab;
 int debug_mode = 2, min_count = 5, min_reduce = 1;
 long long vocab_max_size = 1000, vocab_size = 0;
 long long *vocab_hash;
 long long train_words = 0, file_size = 0;
 
-// Reads a single word from a file, assuming space + tab + EOL to be word boundaries
-void ReadWord(char *word, FILE *fin) {
-  int a = 0, ch;
-  while (!feof(fin)) {
-    ch = fgetc(fin);
-    if (ch == 13) continue;
-    if ((ch == ' ') || (ch == '\t') || (ch == '\n')) {
-      if (a > 0) {
-        if (ch == '\n') ungetc(ch, fin);
-        break;
-      }
-      if (ch == '\n') {
-        strcpy(word, (char *)"</s>");
-        return;
-      } else continue;
-    }
-    word[a] = ch;
-    a++;
-    if (a >= MAX_STRING - 1) a--;   // Truncate too long words
-  }
-  word[a] = 0;
-}
-
-// Returns hash value of a word
-unsigned long long GetWordHash(char *word) {
-  unsigned long long a, hash = 0;
-  for (a = 0; a < strlen(word); a++) hash = hash * 257 + word[a];
-  hash = hash % vocab_hash_size;
-  return hash;
-}
-
-// Returns position of a word in the vocabulary; if the word is not found, returns -1
-unsigned long long SearchVocab(char *word) {
-  unsigned long long hash = GetWordHash(word);
-  while (1) {
-    if (vocab_hash[hash] == -1) return -1;
-    if (!strcmp(word, vocab[vocab_hash[hash]].word)) return vocab_hash[hash];
-    hash = (hash + 1) % vocab_hash_size;
-  }
-  return -1;
-}
-
-// Reads a word and returns its index in the vocabulary
-unsigned long long ReadWordIndex(FILE *fin) {
-  char word[MAX_STRING];
-  ReadWord(word, fin);
-  if (feof(fin)) return -1;
-  return SearchVocab(word);
-}
-
-// Adds a word to the vocabulary
-unsigned long long AddWordToVocab(char *word) {
-  unsigned long long hash, length = strlen(word) + 1;
-  if (length > MAX_STRING) length = MAX_STRING;
-  vocab[vocab_size].word = (char *)calloc(length, sizeof(char));
-  strcpy(vocab[vocab_size].word, word);
-  vocab[vocab_size].cn = 0;
-  vocab_size++;
-  // Reallocate memory if needed
-  if (vocab_size + 2 >= vocab_max_size) {
-    vocab_max_size += 1000;
-    vocab = (struct vocab_word *)realloc(vocab, vocab_max_size * sizeof(struct vocab_word));
-  }
-  hash = GetWordHash(word);
-  while (vocab_hash[hash] != -1) hash = (hash + 1) % vocab_hash_size;
-  vocab_hash[hash] = vocab_size - 1;
-  return vocab_size - 1;
-}
-
-// Used later for sorting by word counts
-int VocabCompare(const void *a, const void *b) {
-    return (int)(((struct vocab_word *)b)->cn - ((struct vocab_word *)a)->cn);
-}
-
-// Sorts the vocabulary by frequency using word counts
-void SortVocab() {
-  long long a, size;
-  unsigned long long hash;
-  // Sort the vocabulary and keep </s> at the first position
-  qsort(&vocab[1], vocab_size - 1, sizeof(struct vocab_word), VocabCompare);
-  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
-  size = vocab_size;
-  train_words = 0;
-  for (a = 0; a < size; a++) {
-    // Words occuring less than min_count times will be discarded from the vocab
-    if ((vocab[a].cn < min_count) && (a != 0)) {
-      vocab_size--;
-      free(vocab[a].word);
-    } else {
-      // Hash will be re-computed, as after the sorting it is not actual
-      hash=GetWordHash(vocab[a].word);
-      while (vocab_hash[hash] != -1) hash = (hash + 1) % vocab_hash_size;
-      vocab_hash[hash] = a;
-      train_words += vocab[a].cn;
-    }
-  }
-  vocab = (struct vocab_word *)realloc(vocab, (vocab_size + 1) * sizeof(struct vocab_word));
-  // Allocate memory for the binary tree construction
-  for (a = 0; a < vocab_size; a++) {
-    vocab[a].code = (char *)calloc(MAX_CODE_LENGTH, sizeof(char));
-    vocab[a].point = (int *)calloc(MAX_CODE_LENGTH, sizeof(int));
-  }
-}
-
-// Reduces the vocabulary by removing infrequent tokens
-void ReduceVocab() {
-  int a, b = 0;
-  unsigned long long hash;
-  for (a = 0; a < vocab_size; a++) if (vocab[a].cn > min_reduce) {
-    vocab[b].cn = vocab[a].cn;
-    vocab[b].word = vocab[a].word;
-    b++;
-  } else free(vocab[a].word);
-  vocab_size = b;
-  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
-  for (a = 0; a < vocab_size; a++) {
-    // Hash will be re-computed, as it is not actual
-    hash = GetWordHash(vocab[a].word);
-    while (vocab_hash[hash] != -1) hash = (hash + 1) % vocab_hash_size;
-    vocab_hash[hash] = a;
-  }
-  fflush(stdout);
-  min_reduce++;
-}
-
-// Learn vocabulary data from file
-void LearnVocabFromTrainFile() {
-  char word[MAX_STRING];
-  FILE *fin;
-  long long a, i;
-  for (a = 0; a < vocab_hash_size; a++) vocab_hash[a] = -1;
-  fin = fopen(train_file, "rb");
-  if (fin == NULL) {
-    printf("ERROR: training data file not found!\n");
-    exit(1);
-  }
-  vocab_size = 0;
-  AddWordToVocab((char *)"</s>");
-  while (1) {
-    ReadWord(word, fin);
-    if (feof(fin)) break;
-    train_words++;
-    if ((debug_mode > 1) && (train_words % 100000 == 0)) {
-      printf("%lldK%c", train_words / 1000, 13);
-      fflush(stdout);
-    }
-    i = SearchVocab(word);
-    if (i == -1) {
-      a = AddWordToVocab(word);
-      vocab[a].cn = 1;
-    } else vocab[i].cn++;
-    if (vocab_size > vocab_hash_size * 0.7) ReduceVocab();
-  }
-  SortVocab();
-  if (debug_mode > 0) {
-    printf("Vocab size: %lld\n", vocab_size);
-    printf("Words in train file: %lld\n", train_words);
-  }
-  file_size = ftell(fin);
-  fclose(fin);
-}
-
 void SaveVocab(char* save_vocab_file) {
   long long i;
   FILE *fo = fopen(save_vocab_file, "wb");

data/ext/word2vec/word2vec.c

@@ -141,6 +141,22 @@ static VALUE build_vocab(VALUE mod, VALUE rb_train_file_name, VALUE rb_vocab_fil
   return Qtrue;
 }
 
+/*
+ * tokenize a file
+ * @param [String] rb_train_file_name 
+ * @param [String] rb_vocab_file_name 
+ * @param [String] rb_output_file_name 
+ */
+static VALUE tokenize(VALUE mod, VALUE rb_train_file_name, VALUE rb_vocab_file_name, VALUE rb_output_file_name) {
+  char* train_filename = StringValueCStr(rb_train_file_name);
+  char* vocab_filename = StringValueCStr(rb_vocab_file_name);
+  char* output_filename = StringValueCStr(rb_output_file_name);
+
+  word2vec_tokenize(train_filename, vocab_filename, output_filename);
+
+  return Qtrue;
+}
+
 void Init_word2vec(void) {
   VALUE mWord2vec = rb_define_module("Word2vec");
   VALUE mWord2vecModel = rb_define_class_under(mWord2vec, "Model", rb_cObject);
@@ -151,4 +167,5 @@ void Init_word2vec(void) {
   rb_define_method(mWord2vecModel, "analogy", model_analogy, 3);
   rb_define_method(mWord2vecModel, "accuracy", model_accuracy, 1);
   rb_define_singleton_method(mWord2vecModel, "build_vocab", build_vocab, 2);
+  rb_define_singleton_method(mWord2vecModel, "tokenize", tokenize, 3);
 }

data/lib/word2vec/version.rb

@@ -1,3 +1,3 @@
 module Word2vec
-    VERSION = "0.4.0"
+    VERSION = "0.5.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: word2vec-rb
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.0
 platform: ruby
 authors:
 - Dani Vela
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2021-06-14 00:00:00.000000000 Z
+date: 2021-06-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -80,6 +80,8 @@ files:
 - ".gitignore"
 - ".rspec"
 - ".travis.yml"
+- ".vscode/c_cpp_properties.json"
+- ".vscode/settings.json"
 - CHANGELOG
 - Gemfile
 - Gemfile.lock
@@ -96,6 +98,7 @@ files:
 - ext/word2vec/common.h
 - ext/word2vec/distance.c
 - ext/word2vec/extconf.rb
+- ext/word2vec/tokenizer.c
 - ext/word2vec/vocab.c
 - ext/word2vec/word2vec.c
 - lib/word2vec.rb