Intro to DS, HW #3 – Map Reduce

Submitted by Ron Kabas (311504484) and Eitan-Hai Mashiah (206349045).

In each section, run the client first from the appropriate folder (using the command in that section), and then in another terminal from the same folder, run the server with the following command:

$ python mincemeat.py -p pass localhost

K-Means

The program loads the points from the points.txt file (located in the k_means folder). It performs a few MapReduce iterations, and stops when none of the centers are updated, which means that all the points picked the same center in that iteration. Finally, it prints the resulting clusters alongside their centers.

To run the client:

$ python k_means.py [-h] [-k K]

where K is the number of required clusters (default: 3). In addition, note that each iteration requires another run of the server.

FindTriangles

The program loads the graph represented by its adjacency list from the grpah.txt file (located in the find_triangle folder), and prints all the triangles (cliques of size 3) in the graph.

To run the client:

$ python find_triangles.py

Pseudo synonyms

The program loads the 3-word queries from the queries.txt file (located in the pseudo_synonyms folder), and prints all the pseudo-synonyms from them.

To run the client:

$ python find_triangles.py

Note that since this task requires two MapReduce stages, the server needs to be run twice sequentially.