AxoyTO

import re
import pandas as pd


def parse_data(files=["1/1_unoptimized.out", "1/1_optimized.out"]):
    threads_list = []
    times_list = []

    for filename in files:
        tmp_threads_list = []

AxoyTO

from sys import argv, exit
import os

MAX_THREADS = 32

def compile(codedir, flag):
    if flag == 'u':
          codefile = codedir + '_unoptimized'
    elif flag == 'o':
        codefile = codedir + '_optimized'

AxoyTO

import pandas as pd
import re


def delete_unnecessary_lines():
    """
    Deleting all other lines, leaves only the last launch results in the file.
    """
    with open("result.out", "r+") as f:
        fpos, seek_pos = 0, 0

AxoyTO

import re
import pandas as pd


def parse_data(filename="result.out"):
    processes_list = []
    sizes_list = []
    T1_times_list = []
    T2_times_list = []
    total_times_list = []

AxoyTO

SHELL = /bin/bash # -> IBM POLUS
#SHELL = /bin/sh # -> MY COMPUTER
SRC = main.cpp functions.cpp
BIN = main
C++FLAGS = -o $(BIN) -std=c++11
C++ = mpic++
#C++POLUS = xlc++ | mpixlc
DATE = $(shell date +%d/%m/%y)
POLUS_MPI_SOURCE = $(shell . mpi_source.sh)

AxoyTO

#!/bin/bash

source /polusfs/setenv/setup.SMPI
make -f Makefile || exit

AxoyTO

from sys import argv, exit
import os
import re


def make_script(
    num_processes_list=[1, 2, 4, 8, 16, 20, 32],
    sizes_list=[128, 256, 512, 1024, 2048, 4096],
):
    with open("mpijob.lsf", "w") as f:

AxoyTO

import re
import pandas as pd


def parse_data(filename="result.out"):
    threads_list = []
    sizes_list = []
    T1_times_list = []
    T2_times_list = []
    total_times_list = []

AxoyTO

from sys import argv, exit
import os
import re


def make_script(num_threads_list=[1, 2, 4, 8, 16, 20, 32, 64],
                sizes_list=[128, 256, 512, 1024]
                ):

    with open('ompjob.lsf', 'w') as f:

AxoyTO

def hausdorff(model_1_vertices, model_2_vertices):
    model_1_vertices = np.array(model_1_vertices[:-1]).reshape(-1, 1)
    model_2_vertices = np.array(model_2_vertices[:-1]).reshape(-1, 1)
    dist = cdist(model_1_points, model_2_points)
    hausdorff_dist = max(dist.min(axis=0).max(), dist.min(axis=1).max())
    return hausdorff_dist