[PERFORMANCE] Adds B_dic for the first word

This design can better train the model and have a better prediction accuracy.
nlpinaction · Dec 10, 2020 · 36aabd3 · 36aabd3
1 parent 8077641
commit 36aabd3
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/chapter-3/分词.ipynb b/chapter-3/分词.ipynb
@@ -177,8 +177,8 @@
     "                        self.Pi_dic[v] += 1  # 每个句子的第一个字的状态，用于计算初始状态概率\n",
     "                    else:\n",
     "                        self.A_dic[line_state[k - 1]][v] += 1  # 计算转移概率\n",
-    "                        self.B_dic[line_state[k]][word_list[k]] = \\\n",
-    "                            self.B_dic[line_state[k]].get(word_list[k], 0) + 1.0  # 计算发射概率\n",
+    "                    self.B_dic[line_state[k]][word_list[k]] = \\\n",
+    "                        self.B_dic[line_state[k]].get(word_list[k], 0) + 1.0  # 计算发射概率\n",
     "        \n",
     "        self.Pi_dic = {k: v * 1.0 / line_num for k, v in self.Pi_dic.items()}\n",
     "        self.A_dic = {k: {k1: v1 / Count_dic[k] for k1, v1 in v.items()}\n",