Merge branch 'reptile-ray'

* reptile-ray: fix typo

Author: alok

reptile/README.md

@@ -3,9 +3,9 @@ PyTorch Implementation of OpenAI's
 but a bit cleaner than John's
 [gist](https://gist.github.com/joschu/f503500cda64f2ce87c8288906b09e2d#file-reptile-sinewaves-demo-py).
 
-It trains 30,00 meta-iterations of gradient descent, one task at a time.
-It evaluates on a fixed task every 1,000 iterations, taking 5 gradient
-descent steps per task. Turns out that 1 is enough to get good
+It trains 30,000 meta-iterations of gradient descent, one task at a
+time. It evaluates on a fixed task every 1,000 iterations, taking 5
+gradient descent steps per task. Turns out that 1 is enough to get good
 performance, indicating that meta-learning is actually working.
 
 ## Requirements
@@ -18,4 +18,4 @@ performance, indicating that meta-learning is actually working.
 
  python3 main.py
 
-It will pop up a matplotlib window that updates every 3,000 iterations.
+It will pop up a Matplotlib window that updates every 3,000 iterations.