Unicity-effectiveness-identification

Here, the code for unicity effectiveness calculation are written in several languages.

Go to each language specific folder to see the code.

The runtime for bigger data (available in Data Generator), is as follows:

SAS:

Average runtime (system): No data yet!

Python :

Average Runtime (System):

For dataset 1: (0.068 + 0.066 + 0.070 + 0.058 + 0.062)/5 = 0.065s

For dataset 2: (0.072 + 0.070 + 0.078 + 0.080 + 0.078)/5 = 0.076s

R:

Average Runtime (system):

For dataset 1: (0.080 + 0.084 + 0.066 + 0.076 + 0.080)/5 = 0.078 s.

For dataset 2: (0.10 + 0.12 + 0.10 + 0.11 + 0.11)/5 = 0.11 s.

Matlab:

Average Runtime:

For dataset 1: (0.0228 + 0.0226 + 0.0230 + 0.0228 + 0.0232)/5 = 0.0229s

For dataset 2: (0.0782 + 0.082 + 0.082 + 0.080 + 0.0832)/5 = 0.0811s

Julia:

Average Runtime (excluding installing packages for the first time and printing the result):

For dataset 1: (2.669322515 + 2.587425153 + 2.587662671 + 2.714858562 + 2.725893324)/5 = 2.657032445

For dataset 2: (2.805421131 + 2.762209794 + 2.721966287 + 2.808386271 + 2.840796785)/5 = 2.7877560536

C:

Average runtime (system):

For dataset 1: (0.016 + 0.015 + 0.016 + 0.014 + 0.012)/5 = 0.015s

For dataset 2: (0.028 + 0.026 + 0.026 + 0.031 + 0.028)/5 = 0.028s

C++:

Average runtime (system):

For dataset 1: (0.035 + 0.033 + 0.033 + 0.036 + 0.035) = 0.035s

For dataset 2: (0.072 + 0.070 + 0.076 + 0.072 + 0.076) = 0.073s

---

The sample data used is as below:

11111 1 1 1 1

12222 1 2 2 2

13333 2 1 3 3

14444 2 2 1 4

15555 3 1 2 1

16666 3 2 3 2

17777 1 3 1 3

18888 2 3 2 4

19999 3 3 3 1

20000 1 1 1 2

21111 2 2 2 3

22222 3 3 3 4

22223 4 1 1 1

22224 4 2 2 2

22225 4 3 3 3

22226 4 4 4 4

The runtimes for each specific languages are as follows:

SAS:

Average runtime (system): (1.74 + 2.4 + 2.8 + 2.85 + 2.93 + 2.84)/6 = 2.59 s.

Python :

Average Runtime (System): (0.063 + 0.074 + 0.042)/3 = 0.059 s.

R:

Average Runtime (system): (0.078 + 0.084 + 0.066)/3 = 0.076 s.

Matlab:

Average Runtime: (0.0218 + 0.0225 + 0.0220)/3 = 0.0221 s.

Julia:

Average Runtime (excluding installing packages for the first time and printing the result): (2.73 + 2.71 + 2.78)/3 = 2.74 s.

Note: Julia seems to be pretty bad in reading input from file because even just such a small file takes around 2 sec to load as a data frame in the script, which pushes the runtime of the code to 2.8 s.

C:

The work hasn't been completed yet. This is taking longer than expected time to debug. Work still in progress. Will update this after we are done.

C++:

Average runtime (system): (0.001 + 0.0009 + 0.001)/3 = 0.00096666666 s.

As we can see, these codes (except Julia for some reason) runs way faster than SAS.