def save_datas(energies_file_pegas, energies_file_dwave, test_set_file, n):
with open(energies_file_pegas, "r") as intput:
energies_pegas = eval(intput.read())
with open(energies_file_dwave, "r") as intput:
energies_dwave = eval(intput.read())
with open(test_set_file, "r") as inf:
dict_from_file = eval(inf.read())
for energy_pegas, energy_dwave, cnf, sat in zip(energies_pegas,
energies_dwave,
dict_from_file['cnf'],
dict_from_file['sat']):
m = len(cnf)
if m + energy_pegas == 0:
en_p = 1
else:
en_p = 0
if m + energy_dwave == 0:
en_dw = 1
else:
en_dw = 0
save_to_csv('max_sat_pegas_vs_2000', [[m, m / n, en_p, en_dw, sat]], n)
def report_random_forest():
size_k = len(list_k)
size_min_samp = len(min_samp_list)
print "k=", size_k, "min_samples=", size_min_samp
matrix = [[0 for x in range(size_min_samp)] for y in range(size_k)]
i = 0
j = 0
last_value_min_samples = list_k[0]
for param, mean_score, cv_scores in clf.grid_scores_:
if param["feature_selection__k"] != last_value_min_samples:
i += 1
j = 0
last_value_min_samples = param["feature_selection__k"]
#print param, "mean=", mean_score, "i=", i, "j=", j
matrix[i][j] = mean_score
j += 1
import numpy as np
header = [""] + min_samp_list
matrix = np.c_[list_k, matrix]
u.save_to_csv("RandomForest.csv", Algorithms_path, matrix, header)
u.change_decimal_separator("RandomForest.csv", Algorithms_path)
def report_decision_tree():
size_k = len(list_k)
size_min_samp = len(min_samp_list)
print "k=", size_k, "min_samples=", size_min_samp
matrix = [[0 for x in range(size_min_samp)] for y in range(size_k)]
i = 0
j = 0
last_value_min_samples = min_samp_list[0]
for param, mean_score, cv_scores in clf.grid_scores_:
if param[
"decision_tree__min_samples_split"] != last_value_min_samples:
i += 1
j = 0
last_value_min_samples = param[
"decision_tree__min_samples_split"]
#print param, "mean=", mean_score, "i=", i, "j=", j
matrix[j][i] = mean_score
j += 1
import numpy as np
header = [""] + min_samp_list
matrix = np.c_[list_k, matrix]
u.save_to_csv("DecisionTree.csv", Algorithms_path, matrix, header)
u.change_decimal_separator("DecisionTree.csv", Algorithms_path)
def report_SVM():
size_k = len(list_k)
size_c = len(list_C)
print "k=", size_k, "c=", size_c
matrix = [[0 for x in range(size_c)] for y in range(size_k)]
i = 0
j = 0
last_value_C = list_C[0]
for param, mean_score, cv_scores in clf.grid_scores_:
if param["SVM__C"] != last_value_C:
i += 1
j = 0
last_value_C = param["SVM__C"]
#print param, "mean=", mean_score, "i=", i, "j=", j
matrix[j][i] = mean_score
j += 1
import numpy as np
header = [""] + list_C
matrix = np.c_[list_k, matrix]
u.save_to_csv("SVM.csv", Algorithms_path, matrix, header)
u.change_decimal_separator("SVM.csv", Algorithms_path)
def pegasus_phase_draw(energies_file, test_set_file, n):
with open(energies_file, "r") as intput:
energies = eval(intput.read())
with open(test_set_file, "r") as inf:
dict_from_file = eval(inf.read())
for energy, cnf , sat in zip(energies, dict_from_file['cnf'], dict_from_file['sat']):
m = len(cnf)
if m + energy == 0:
en = 1
else:
en = 0
save_to_csv('max_sat_', [[m, m / n, en , sat]], n)
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
df = pd.read_csv('logs_max_sat__{0}.csv'.format(n), names=['M','alfa','y_hat','y_acc'], sep=',').sort_values(['alfa'])[0:1000]
bins = np.arange(0, 2.1, 0.20)
df['bins'] = pd.cut(df.alfa, bins)
ss_acc = df.groupby('bins')['y_acc'].agg(['sum', 'count'])
ss_hat = df.groupby('bins')['y_hat'].agg(['sum', 'count'])
y = [x.mid for x in ss_acc.index.values]
plt.title('2-SAT, each point is averaged by step: (1/20)', fontsize=14)
plt.plot(y, ss_hat['sum']/ss_hat['count'], "-",label='Pegasus-{0}'.format(n))
plt.plot(y, ss_acc['sum']/ss_acc['count'], "-",label='MiniSAT-{0}'.format(n))
plt.xticks(fontsize = 15)
plt.yticks(fontsize = 15)
plt.grid()
plt.legend(fontsize=12)
plt.show()
plt.savefig('phase{0}'.format(n))
plt.close()
def main():
# for logging purposes
start_time = time()
# create the sitemap and convert the dictionaries to CSV
result = create_sitemap(START_URL, EXPLORED, TO_VISIT)
sitemap, n_pages = result[0], result[1]
to_csv = dict_to_csv(sitemap)
elapsed_time = time() - start_time
stats = f"Explored { n_pages - 1 } pages in { elapsed_time } seconds"
print(colored("=" * 50, "white"))
print(colored(stats, "blue"))
# write the data to a sitemap CSV file (easily exportable to Excel)
save_to_csv(to_csv, "sitemap")
print(colored("Successfully saved sitemap to CSV", "magenta"))