Move Numba example to its own directory

Author: gjbex

source-code/README.md

@@ -10,8 +10,6 @@ Sample code for performing computations on a GPU.
    numbers on a GPU.
 1. `scikit_cuda.ipynb`: Jupyter notebook illustrating linear algebra
    on a GPU device.
-1. `numba.ipynb`: Jupyter notebook illustrating using numba for
-   GPU computing.
-
+1. `numba`: code illustrating using numba for GPU computing.
 1. `cupy.ipynb`: Jupyter notebook illustrating some aspects of
    the `cupy` package.

source-code/cupy.ipynb renamed to source-code/cupy/cupy.ipynb

@@ -0,0 +1,11 @@
+# Numba
+
+Numba can be used to compile Python code to run on GPUs.  This directory has
+code samples to illustrate this.
+
+
+## What is it?
+
+1. `numba.ipynb`: Jupyter notebook illustrating how to use Numba to compile
+    Python code to run on GPUs.  It illustrates the typical pitfalls for GPU
+    programming (e.g., memory access patterns), and how to avoid them.