Update 4_cifar10_tutorial.ipynb

hipparic · Feb 19, 2019 · eb895c6 · eb895c6
1 parent 57f3b5d
commit eb895c6
Showing 1 changed file with 4 additions and 6 deletions.
diff --git a/chapter1/4_cifar10_tutorial.ipynb b/chapter1/4_cifar10_tutorial.ipynb
@@ -60,9 +60,8 @@
     "\n",
     "\n",
     "1. 读取和归一化 CIFAR10\n",
-    "----------------------------\n",
     "\n",
-    "使用``torchvision``可以非常容易的加载CIFAR10\n"
+    "使用``torchvision``可以非常容易的加载CIFAR10。\n"
    ]
   },
   {
@@ -80,7 +79,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "torchvision的输出是[0,1]的PILImage图像,我们把它转换为归一化范围为[-1, 1]的张量.\n"
+    "torchvision的输出是[0,1]的PILImage图像，我们把它转换为归一化范围为[-1, 1]的张量。\n"
    ]
   },
   {
@@ -133,7 +132,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "我们展示一些训练图像.for fun\n",
+    "我们展示一些训练图像。\n",
     "\n"
    ]
   },
@@ -185,9 +184,8 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "2. 定义一个卷积神经网络\n",
     "\n",
-    "^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
+    "2. 定义一个卷积神经网络\n",
     "\n",
     "从之前的神经网络一节复制神经网络代码,并修改为输入3通道图像\n"
    ]