Update 5_data_parallel_tutorial.ipynb

hipparic · Feb 19, 2019 · 9092b48 · 9092b48
1 parent e2b5e87
commit 9092b48
Showing 1 changed file with 4 additions and 4 deletions.
diff --git a/chapter1/5_data_parallel_tutorial.ipynb b/chapter1/5_data_parallel_tutorial.ipynb
@@ -36,7 +36,7 @@
     "请注意，只调用``my_tensor.to(device)``并没有复制张量到GPU上，而是返回了一个copy。所以你需要把它赋值给一个新的张量并在GPU上使用这个张量。\n",
     "\n",
     "在多GPU上执行前向和反向传播是自然而然的事。\n",
-    "但是PyTorch默认将只是用一个GPU。\n",
+    "但是PyTorch默认将只使用一个GPU。\n",
     "\n",
     "使用``DataParallel``可以轻易的让模型并行运行在多个GPU上。\n",
     "\n",
@@ -139,7 +139,7 @@
     "简单模型\n",
     "------------\n",
     "作为演示，我们的模型只接受一个输入，执行一个线性操作，然后得到结果。\n",
-    "说明，``DataParallel``能在任何模型（CNN，RNN，Capsule Net等）上使用。\n",
+    "说明：``DataParallel``能在任何模型（CNN，RNN，Capsule Net等）上使用。\n",
     "\n",
     "\n",
     "我们在模型内部放置了一条打印语句来打印输入和输出向量的大小。\n",
@@ -286,7 +286,7 @@
     "    Outside: input size torch.Size([10, 5]) output_size torch.Size([10, 2])\n",
     "\n",
     "3 GPUs\n",
-    "~~~~~~\n",
+    "~\n",
     "\n",
     "If you have 3 GPUs, you will see:\n",
     "\n",
@@ -311,7 +311,7 @@
     "    Outside: input size torch.Size([10, 5]) output_size torch.Size([10, 2])\n",
     "\n",
     "8 GPUs\n",
-    "~~~~~~~~~~~~~~\n",
+    "~~\n",
     "\n",
     "If you have 8, you will see:\n",
     "\n",