In this project, the goal is to implement a parallel version of the Monte Carlo method for calculating

$\backslash$pi using Numba and CUDA. The idea is that you will simulate taking a random point out of a square

shape as shown in Figure $1.$ If you divide the number of points within the circle to the total number

of points, then you will get an estimate on the area of the circle, which can be used to infer the value

of $\backslash$pi $.$

The report for your project should include your choices of multiple groups of parameters for the

CUDA kernel, by changing the number of blocks and threads. Your result should also show the

latency for the calculation, and demonstrate how the accuracy improves with more samples. You

should include in your report a table showing the correlation of accuracy with the number of

samples, as well as the computational time. The table is illustrated as below. You should include at

least ten rows in the report