Fixed typos

.gitignore

@@ -5,6 +5,7 @@
 *.pt
 X*
 ppts
+.ipynb_checkpoints
 client_secret.json
 __pycache__/
 tmp

tutorials/PyTorchDataParallel.ipynb

@@ -37,16 +37,17 @@
     "```python\n",
     "mytensor = my_tensor.gpu()\n",
     "```\n",
-    "Please note that just calling `mytensor.gpu()` won't copy the tesor to GPU. You need to assign it to a new tensor and use the tensor on GPU.\n",
+    "Please note that just calling `mytensor.gpu()` won't copy the tensor to GPU. You need to assign it to a new tensor and use the tensor on GPU.\n",
     "\n",
-    "Furthermore, it's natural to execute your long-waiting forward, backward propagations on multiple GPUs. Unfortunatly, PyTorch won't do that automatically for you. Not yet. (It will just use one GPU for you.) \n",
+    "Furthermore, it's natural to execute your long-waiting forward, backward propagations on multiple GPUs. Unfortunately, PyTorch won't do that automatically for you. Not yet. (It will just use one GPU for you.) \n",
     "\n",
-    "However, running your operations on multiple GPUs is very easy. Jusy you need to make your model dataparallelable using this.  \n",
+    "However, running your operations on multiple GPUs is very easy. Just you need to make your model dataparallelable using this.  \n",
     "```python\n",
     "model = nn.DataParallel(model)\n",
     "```\n",
     "\n",
-    "That's it. If you want to know more, here we are!\n"
+    "That's it. If you want to know more, here we are!\n",
+    "\n"
    ]
   },
   {
@@ -57,7 +58,7 @@
    },
    "source": [
     "## Imports and parameters\n",
-    "Let's import our favorate core PyTorch things and define some paramters."
+    "Let's import our favorite core PyTorch things and define some parameters. "
    ]
   },
   {
@@ -140,9 +141,9 @@
    },
    "source": [
     "## Simple Model\n",
-    "Then, we need a model to run. For DataParallel demo, let's make a simple one. Just get an input and do a linear operation, and output . However, you can make any model including CNN, RNN or even Capsule Net for `DataParallel`.\n",
+    "Then, we need a model to run. For DataParallel demo, let's make a simple one. Just get an input and do a linear operation, and output. However, you can make any model including CNN, RNN or even Capsule Net for `DataParallel`.\n",
     "\n",
-    "Inside of the model, we just put a print statement to monitor the szie of input and outout tensors. Please pay attention on the batch part, rank 0 when they print out something."
+    "Inside of the model, we just put a print statement to monitor the size of input and output tensors. Please pay attention to the batch part, rank 0 when they print out something."
    ]
   },
   {
@@ -184,7 +185,7 @@
    },
    "source": [
     "## Create Model and DataParallel\n",
-    "Here is the core part. First make a model instance, and check if you have multiple GPUs. (If you don't, I feel sorry for you.) If you have, just wrap our model using `nn.DataParallel`. That's it. I know, it's hard to believe, but that's really it!\n",
+    "Here is the core part. First, make a model instance, and check if you have multiple GPUs. (If you don't, I feel sorry for you.) If you have, just wrap our model using `nn.DataParallel`. That's it. I know, it's hard to believe, but that's really it!\n",
     "\n",
     "Then, finally put your model on GPU by `model.gpu()`. It's simple and beautiful.\n"
    ]
@@ -223,7 +224,7 @@
    },
    "source": [
     "## Fun part\n",
-    "Now it's the fun part. Just get data from the dataloader and see what are the size of input and out tensors!"
+    "Now it's the fun part. Just get data from the dataloader and see the size of input and out tensors!"
    ]
   },
   {
@@ -295,9 +296,9 @@
    },
    "source": [
     "## Didn't you see?\n",
-    "Hmm, did you see something working here? It seems just batch 30 input and outout 30. The model gets 30 and spit out 30. Nothing special.\n",
+    "Hmm, did you see something working here? It seems just batch 30 input and output 30. The model gets 30 and spits out 30. Nothing special.\n",
     "\n",
-    "BUT, Wait! This notebook (or yours) does not have GPUs. If you have GPUs, the execution looks like this, caled DataParallel!\n",
+    "BUT, Wait! This notebook (or yours) does not have GPUs. If you have GPUs, the execution looks like this, called DataParallel!\n",
     "\n",
     "### 2 GPUs\n",
     "```bash\n",