Fixed typo

HT to Anthony Voutas

Author: chewxy

content/post/deeplearning_in_go_part_1.md

@@ -345,7 +345,7 @@ I would however, posit at least three advantages of having a graph object. All o
 
 ### Numerical Stability ###
 
-Consider the equation $y = log(1 + x)$. This equation is not [numerically stable](https://en.wikipedia.org/wiki/Numerical_stability) - for very small values of `x`, the answer will most likely be wrong. This is because of the way `float64` is designed - a `float64` does not have enough bits to be able to tell apart `1` and `1 + 10e-16`. In fact, the correct way to do $latex y = log(1 + x)$ is to use the built in library function `math.Log1p`. It can be shown in this simple program:
+Consider the equation $y = log(1 + x)$. This equation is not [numerically stable](https://en.wikipedia.org/wiki/Numerical_stability) - for very small values of `x`, the answer will most likely be wrong. This is because of the way `float64` is designed - a `float64` does not have enough bits to be able to tell apart `1` and `1 + 10e-16`. In fact, the correct way to do $ y = log(1 + x)$ is to use the built in library function `math.Log1p`. It can be shown in this simple program:
 
 ```go
 func main() {
@@ -401,4 +401,4 @@ If you're working on a truly homoiconic language such as lisp or Julia, you prob
 
 Do note that this isn't a knock on Go or Python or Lua. All of these languages have their strengths and weaknesses. But why do deep learning related work in Go when there are more mature libraries in Python or Lua? Well, one of the major reasons I developed Gorgonia was the ability to deploy everything neatly into one single binary. Doing that with Python or Lua would take an immense amount of effort. By contrast, deploying Go programs are a breeze. 
 
-I believe that Go for data science is an amazing idea. It is typesafe (enough for me), and it's compiled down to binary. Go allows for better mechanical sympathy, which I believe is key to having faster and better AI out there. Afterall, we are ALL bound by our hardware. I just wish there were better higher level data structures for me to express my ideas. There weren't, so I built them. And I hope you use them.
+I believe that Go for data science is an amazing idea. It is typesafe (enough for me), and it's compiled down to binary. Go allows for better mechanical sympathy, which I believe is key to having faster and better AI out there. Afterall, we are ALL bound by our hardware. I just wish there were better higher level data structures for me to express my ideas. There weren't, so I built them. And I hope you use them.