minor spelling tweaks (#1277)

userguide/activity_classifier/how-it-works.md

@@ -12,6 +12,6 @@ The output of the convolutional layer is a vector representation for each predic
 
 The recurrent layer is specialized for learning temporal features across sequences. For example it may learn that spatial features associated with walking are more likely to occur after detecting spatial features associated with running. These features are further encoded into the output of the recurrent layer.
 
-In order to detect these features along sessions the recurrent layer takes into account it's own **state** - the output of the recurrent layer for the previous prediction window. The output of the recurrent layer for the current prediction window is turned into a probability vector across all desired activities to produce the final classification.
+In order to detect these features along sessions the recurrent layer takes into account its own **state** - the output of the recurrent layer for the previous prediction window. The output of the recurrent layer for the current prediction window is turned into a probability vector across all desired activities to produce the final classification.
 
 More about this type of deep learning architecture can be found in [Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition](http://www.mdpi.com/1424-8220/16/1/115).

userguide/clustering/dbscan.md

@@ -157,7 +157,7 @@ ax.scatter(sf['X1.0'][noise_mask], sf['X1.1'][noise_mask], s=40, alpha=0.7,
 fig.show()
 ```
 
-![dbscan_mooons](images/dbscan_moons.png)
+![dbscan_moons](images/dbscan_moons.png)
 
 For comparison, K-means cannot identify the true clusters in this case, even
 when we tell the model the correct number of clusters.
@@ -173,7 +173,7 @@ ax.scatter(sf['X1.0'], sf['X1.1'], s=80, alpha=0.9, c=sf['kmeans_id'],
 fig.show()
 ```
 
-![kmeans_mooons](images/kmeans_moons.png)
+![kmeans_moons](images/kmeans_moons.png)
 
 
 #### Setting key parameters

userguide/cpp.md

@@ -188,7 +188,7 @@ Data:
 To build this C++ program, you'll need to borrow the compilation flags from Turi Create as shown above.
 
 ```shell
-# run the following in the turicreate repostitory root:
+# run the following in the turicreate repository root:
 ./configure
 cd release/src/unity
 make -j8

userguide/image_similarity/README.md

@@ -77,7 +77,7 @@ reference_data[9]['image'].show()
 ```
 ![Similar images](images/sample_image.jpg)
 Note: the loading order of images is non-deterministic, so your 10th image
-is likely something differnt but this isn't important for demonstration purposes.
+is likely something different but this isn't important for demonstration purposes.
 
 The 10 most "similar" looking images are
 ```python

userguide/style_transfer/export-coreml.md

@@ -1,6 +1,6 @@
 # Deployment to Core ML
 
-Style trasnfer models created in Turi Create can easily be deployed to
+Style transfer models created in Turi Create can easily be deployed to
 Core ML into an iOS or macOS application.
 
 #### Export to Core ML {#coreml}
@@ -19,7 +19,7 @@ arrow next to MyCustomStyleTransfer should appear:
 
 ![Xcode view of MyStyleTransfer.mlmodel](images/xcode_style_transfer.png)
 
-If your style transfer model has 8 styles and you want to stlyize using
+If your style transfer model has 8 styles and you want to stylize using
 the 4th style, then index should be
 
 ```swift