def print_averages(df):
averages = df[['accuracy', 'precision', 'recall', 'fscore']].mean(axis=0)
print(indent('\nPrinting averages:', spaces=10))
for label, avg in averages.iteritems():
print(indent('* Average %s: %s' % (label, str(avg)), spaces=14))
def compute_stats(metrics, classifiers, folder):
print(indent('\n- Computing metrics statistics ... '), end='')
stats = pd.DataFrame(columns=classifiers)
for key, val in metrics.items():
name = str(capitalize(key))
mean = val.mean(axis=0)
mean = mean.rename(name+' Mean')
stats = stats.append(mean)
median = val.median(axis=0)
median = median.rename(name+' Median')
stats = stats.append(median)
std = val.std(axis=0)
std = std.rename(name+' Standard Deviation')
stats = stats.append(std)
print('result:')
print(indent(stats.to_string(), spaces=10))
out = write_df_to_csv(folder, stats, 'stats')
print(indent('\n- Statistics written to file "%s"' % out))
def find_best_k_all(fvs, n=MAX_k):
for path, name in fvs:
print('\n> Filename: %s\n' % name)
print(indent('cluster\tK-Means\t Agglomerative'))
[
print(indent('\t\t'.join(m)))
for m in find_best_k(get_fv_values(path), n + 1)
]
def run_cross_validation_all(data, labels):
print(indent('\n- Running cross validation with 5-folds ...'), end='\n\n')
all_prec, all_recall, all_f1 = pd.DataFrame(), pd.DataFrame(), pd.DataFrame()
for label, clf in CLASSIFIERS.items():
print(indent('* Validating "%s"' % label, spaces=10))
df = run_cross_validation(clf, data, labels)
all_prec[label] = df['test_precision']
all_recall[label] = df['test_recall']
all_f1[label] = df['test_f1']
return {
'precision': all_prec,
'recall': all_recall,
'fscore': all_f1
}
def find_god_classes(source=None):
print('\n> Looking for god classes in "%s"...' % os.path.abspath(source))
# creates new dataframe
df = pd.DataFrame(columns=['class_name', 'method_num', 'path_to_source'])
# goes through the project content
for root, dirs, files in os.walk(source, topdown=False):
for name in files:
# mathces only java sources
if name.endswith('.java'):
with(open(root + '/' + name, 'r')) as jsc:
# creates the AST
tree = jl.parse.parse(jsc.read())
# iterates through the AST (first filtering all the non-class declarations)
for path, node in tree.filter(jl.parser.tree.ClassDeclaration):
# adds the new class to the data frame (as well as the class details)
df = df.append({
'class_name': node.name,
'path_to_source': root + '/' + node.name + '.java',
'method_num': len(node.methods)
}, ignore_index=True, sort=-1)
# Filters the dataframe getting only the god classes
god_classes = filter_all_classes(df)
print('> Found the following god classes:\n'+indent(god_classes.to_string()))
return god_classes
def run_wilcoxon_test_all(metrics, folder):
b_folder = mkdir(folder + '/wilcoxon')
for key, val in metrics.items():
run_wilcoxon_test(val, key, b_folder)
print(indent('\n- Wilcoxon tests results written to folder "%s"' % b_folder))
def biased_clf_metrics_to_csv(labels, folder):
prec, rec, fscore, sup = precision_recall_fscore_support(labels, ones(len(labels)), average='binary')
df = pd.DataFrame({'precision': prec, 'recall': rec, 'fscore': fscore}, index=[0])
out = write_df_to_csv(folder, df, 'biased_metrics')
print(indent('\n- Biased classifier metrics ("precision", "recall" and "fscore") written to file "%s"' % out))
def make_boxplot_all(metrics, folder):
b_folder = mkdir(folder + '/boxplots')
for key, val in metrics.items():
make_boxplot(val, key, b_folder)
print(indent('\n- Boxplots written to folder "%s"' % b_folder))
def gen_metrics_biased_clf(data, labels):
print(indent('\n- Generating biased estimators ...'))
l_prec, l_recall, l_fscore = [], [], []
for i in range(0, 100):
prec, rec, f1, sup = split_and_compute_stats(data, labels)
l_prec.append(prec)
l_recall.append(rec)
l_fscore.append(f1)
return l_prec, l_recall, l_fscore
def make_plot(x, data, title, folder):
plt.figure(figsize=(26, 10))
plt.title(title)
plt.xlabel('Run')
plt.ylabel('Metrics')
for col in data.columns:
plt.plot(x, data[col].values, marker='o', markersize=4, label=col)
plt.legend()
plt.grid(True)
path = folder + '/' + title.replace(" ", "") + '.png'
plt.savefig(path)
plt.clf()
print(indent('\nPlot saved to "%s"' % path, spaces=10))
Populates the ontology (tree.owl) with instances of ClassDeclaration, MethodDeclaration, FieldDeclaration,
Statement subclasses (e.g., IfStatement, WhileStatement, etc.) and FormalParameter.
Example usage:
$ python3 bad_smells.py individ_creator
flags:
-s <path-to-source> | --source <path-to-source>:
The path to a directory containing java files. Default is "PROJ_ROOT/lib/android-chess/app/src/main/java/jwtc/chess"
''',
'find_bad_smells': '''
find_bad_smells:
Runs Sparql queries to detect bad smells (Long Method, Large Class, Long Parameter List, Switch Statements, Data Class).
Example usage:
$ python3 bad_smells.py find_bad_smells
'''
}
def docstring_preview(text):
return text.split('\n\n')[0]
docstring_headers = {
key: indent(docstring_preview(value))
for (key, value) in MODULE_DOCSTRINGS.items()
}
DOCSTRING = DOCSTRING.format(**docstring_headers)