revised the way that rhyming words are found. This should spend less …

…time trying words that aren't rhymes at the end of lines.

make_rhyming_tokens.py

@@ -0,0 +1,25 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Thu May 14 20:55:21 2020
+
+@author: Doug
+"""
+
+import pickle
+from collections import defaultdict
+
+rhyming_tokens = defaultdict(set)
+syllable_tokens = [set() for y in range(10)]
+for word in rhyme_dictionary:
+    rhyme_part = rhyme_dictionary[word]
+    rhyming_words = reverse_rhyme_dictionary[rhyme_part]
+    for rhyming_word in rhyming_words:
+        these_rhyming_tokens=tokenizer.encode(rhyming_word,add_prefix_space = True)
+        rhyming_tokens[word] = rhyming_tokens[word].union(set(these_rhyming_tokens))
+    word_tokens=tokenizer.encode(word,add_prefix_space = True)
+    syll_count = min(syllable_count_dictionary[word.upper()],9)
+    syllable_tokens[syll_count] = syllable_tokens[syll_count].union(word_tokens)
+
+with open("rhyming_tokens.p","wb") as f:
+    pickle.dump(rhyming_tokens, f)
+    pickle.dump(syllable_tokens, f)