Fix. Correct the wrong words in the sentence

NLP通用框架BERT项目实战/第一章——NLP通用框架BERT原理解读.md

@@ -206,7 +206,7 @@ Multi-Head架构图如下
 
 - 方法1：句子中有15%的词汇被随机mask掉
 - 较给模型去预测被mask的是什么
-- 词语的可能性太多了，中文一般是子
+- 词语的可能性太多了，中文一般是字
 - 如果BERT训练的向量好，那分类自然好
 
 ![1609751713233](assets/1609751713233.png)

机器学习竞赛实战_优胜解决方案/信用卡欺诈检测/.ipynb_checkpoints/逻辑回归-信用卡欺诈检测-checkpoint.ipynb

@@ -599,7 +599,7 @@
     "random_normal_indices=np.random.choice(normal_indices, number_records_fraud,replace=False) \n",
     "random_normal_indices=np.array(random_normal_indices)\n",
     "\n",
-    "# 有了正常样本和异常样本的所以\n",
+    "# 有了正常样本和异常样本的索引\n",
     "under_sample_indices=np.concatenate([fraud_indices,random_normal_indices])\n",
     "\n",
     "# 根据索引得到下采样的所有样本点\n",

机器学习竞赛实战_优胜解决方案/信用卡欺诈检测/逻辑回归-信用卡欺诈检测.ipynb

@@ -599,7 +599,7 @@
     "random_normal_indices=np.random.choice(normal_indices, number_records_fraud,replace=False) \n",
     "random_normal_indices=np.array(random_normal_indices)\n",
     "\n",
-    "# 有了正常样本和异常样本的所以\n",
+    "# 有了正常样本和异常样本的索引\n",
     "under_sample_indices=np.concatenate([fraud_indices,random_normal_indices])\n",
     "\n",
     "# 根据索引得到下采样的所有样本点\n",

机器学习竞赛实战_优胜解决方案/机器学习实战小项目/文本特征处理方法对比/NLP处理实例.ipynb

@@ -1528,7 +1528,7 @@
    "source": [
     "cm_w2v = confusion_matrix(y_test_word2vec, y_predicted_word2vec)\n",
     "fig = plt.figure(figsize=(10, 10))\n",
-    "plot = plot_confusion_matrix(cm, classes=['Irrelevant','Disaster','Unsure'], normalize=False, title='Confusion matrix')\n",
+    "plot = plot_confusion_matrix(cm_w2v, classes=['Irrelevant','Disaster','Unsure'], normalize=False, title='Confusion matrix')\n",
     "plt.show()\n",
     "print(\"Word2Vec confusion matrix\")\n",
     "print(cm_w2v)\n",