diff --git a/机器学习竞赛实战_优胜解决方案/快手短视频用户活跃度分析/.ipynb_checkpoints/快手用户活跃预测-checkpoint.ipynb b/机器学习竞赛实战_优胜解决方案/快手短视频用户活跃度分析/.ipynb_checkpoints/快手用户活跃预测-checkpoint.ipynb
index f665094..e420f6b 100644
--- a/机器学习竞赛实战_优胜解决方案/快手短视频用户活跃度分析/.ipynb_checkpoints/快手用户活跃预测-checkpoint.ipynb
+++ b/机器学习竞赛实战_优胜解决方案/快手短视频用户活跃度分析/.ipynb_checkpoints/快手用户活跃预测-checkpoint.ipynb
@@ -49225,6 +49225,8 @@
     }
    ],
    "source": [
+    "# tf.get_variable创建新的tensorflow变量\n",
+    "# tf.placeholder在神经网络构建graph的时候在模型中的占位，此时没有把数据传入模型，只会分配必要的内存。运行模型的时候通过feed_dict()函数向占位符喂入数据。\n",
     "n_features = 12\n",
     "n_hu = 8\n",
     "with tf.variable_scope('train'):     # tf.variable_scope用来指定变量的作用域\n",
@@ -49233,7 +49235,7 @@
     "    lr = tf.placeholder(tf.float32, [], name='learning_rate')    # 定义学习率\n",
     "    \n",
     "    # 隐藏层到输出层的参数W, b，n_hu隐藏单元的个数\n",
-    "    W_out = tf.get_variable('W_out', [n_hu, 1])   \n",
+    "    W_out = tf.get_variable('W_out', [n_hu, 1])\n",
     "    b_out = tf.get_variable('b_out', [1])\n",
     "    \n",
     "    # x(batch_size, seq_length, n_features)\n",