def explore_scatter(self, X):
feature_cols = [
'age', 'job', 'marital', 'education', 'default', 'loan', 'housing',
'contact', 'poutcome'
]
age = X['age']
job = X['job']
marital = X['marital']
education = X['education']
housing = X['housing']
contact = X['contact']
loan = X['loan']
plot_scatter(age, job, 'Age', 'Job', 'Age vs Job')
plot_scatter(marital, education, 'marital', 'education',
'marital vs education')
plot_scatter(housing, contact, 'housing', 'contact',
'housing vs contact')
plot_scatter(loan, education, 'loan', 'education', 'loan vs education')
plot_scatter(job, marital, 'job', 'marital', 'job vs marital')
def explore_scatter(self):
feature_cols = [
"fixed acidity", "residual sugar", "chlorides",
"total sulfur dioxide", "density", "alcohol"
]
fa = self.data['fixed acidity']
rs = self.data['residual sugar']
c = self.data['chlorides']
tso = self.data['total sulfur dioxide']
d = self.data['density']
a = self.data['alcohol']
plot_scatter(fa, rs, 'Fixed Acidity', 'Residual Sugar',
'Fixed Acidity vs Residual Sugar')
plot_scatter(c, tso, 'Chlorides', 'Total Sulfur Dioxide',
'Chlorides vs total sulfur dioxide')
plot_scatter(d, a, 'Density', 'Alcohol', 'Density vs Alcohol')
plot_scatter(fa, a, 'Fixed Acidity', 'Alcohol',
'Fixed Acidity vs Alcohol')
plot_scatter(rs, a, 'Residual Sugar', 'Alcohol',
'Residual Sugar vs Alcohol')
def main():
options = parser.parse_args()
inputFile = options.inputFile
df = pd.read_csv(inputFile, sep="|", index_col=COLUMNS[0])
#check_limits(df)
duplicateRowsDF = df[df.duplicated()]
if len(duplicateRowsDF.index) > 0:
print("Warning, there are some duplicates")
points = np.column_stack([df[COLUMNS[2]].values, df[COLUMNS[3]].values])
pareto_front = df.iloc[is_pareto_efficient(points, return_mask=False)]
out_name = options.outputFile.split('.')[0]
if options.render:
plotter.plot_scatter(df[COLUMNS[3]].values,
df[COLUMNS[2]].values,
pareto_front=(pareto_front[COLUMNS[3]].values,
pareto_front[COLUMNS[2]].values),
label=str(options.mrai_type) +
" MRAI experiments",
xlabel="Messages transmitted",
ylabel="Convergence time [s]",
title="Experiments efficency",
output_file_name=out_name + ".pdf")
allowed_types = ["constant", "dpc", "dpc2", "reverse_dpcc"]
if options.mrai_type in allowed_types:
plotter.plot_messages_time_comparison(
df[COLUMNS[1]].values,
df[COLUMNS[2]].values,
df[COLUMNS[3]].values,
title="MRAI " + options.mrai_type + " performances",
output_file_name=out_name + "_mrai_evolution.pdf",
time_max=options.time_max,
time_min=options.time_min,
messages_max=options.messages_max,
messages_min=options.messages_min,
fs=options.font_size)
plotter.plot_suppression(df[COLUMNS[1]].values,
df[COLUMNS[4]].values,
title="Suppression detected",
output_file_name=out_name +
"_suppression_evolution.pdf",
sup_max=options.rfd_max,
sup_min=options.rfd_min,
fs=options.font_size)
plotter.plot_messages_suppression_time_comparison(
df[COLUMNS[1]].values,
df[COLUMNS[2]].values,
df[COLUMNS[3]].values,
df[COLUMNS[4]].values,
title="MRAI + RFD " + options.mrai_type + " performances",
output_file_name=out_name + "_mrai_rfd_evolution.pdf",
time_max=options.time_max,
time_min=options.time_min,
messages_max=options.messages_max,
messages_min=options.messages_min,
rfd_max=options.rfd_max,
rfd_min=options.rfd_min,
fs=options.font_size)
plotter.plot_messages_time_comparison_error_bars(
df[COLUMNS[1]].values,
df[COLUMNS[2]].values,
df[COLUMNS[3]].values,
df[COLUMNS[6]].values,
df[COLUMNS[9]].values,
title="MRAI " + options.mrai_type + " performances " +
"with Std deviation",
output_file_name=out_name + "_mrai_evolution_std.pdf",
time_max=options.time_max,
time_min=options.time_min,
messages_max=options.messages_max,
messages_min=options.messages_min,
fs=options.font_size)
plotter.plot_messages_time_comparison_error_bars_alpha(
df[COLUMNS[1]].values,
df[COLUMNS[2]].values,
df[COLUMNS[3]].values,
df[COLUMNS[6]].values,
df[COLUMNS[9]].values,
title="MRAI " + options.mrai_type + " performances " +
"with Std deviation",
output_file_name=out_name + "_mrai_evolution_std_alpha.pdf",
time_max=options.time_max,
time_min=options.time_min,
messages_max=options.messages_max,
messages_min=options.messages_min,
fs=options.font_size)
sys.path.append('plotter/plotting_params')
params_module = importlib.import_module(args.params_file)
data_params = params_module.data_params
if hasattr(params_module, 'list_of_experiments'):
final_list_of_experiments = params_module.list_of_experiments
else:
final_list_of_experiments = list_of_experiments
print('final_list_experiments', final_list_of_experiments)
print('data params', data_params)
print('process params', params_module.processing_params)
print('plot params', params_module.plot_params)
if args.erase_bad_validations:
validations = [
'Town01W1Noise', 'Town02W14Noise', 'Town01W1', 'Town02W14'
]
erase_wrong_plotting_summaries(args.folder, validations)
#if check_csv_ground_truths
# Check if the validation folders already have the
plot_scatter(args.folder,
final_list_of_experiments,
data_params,
params_module.processing_params,
params_module.plot_params,
out_folder=args.params_file)