f1scores = [[[] for i in range(len(test_channels))] for j in range(len(train_channels))]
for run in range(run_nr):
print('run ' + str(run))
# firstly split the input into train test
doas_train, doas_test, ind_test = train_test_doa(doas, 0.2)
for ind_train, ch_train in enumerate(train_channels):
for ind_test, ch_test in enumerate(test_channels):
print("start running for channel " + str(ch_train) + ' and ' + str(ch_test) + ' ' + segment + '\n')
# SplitData(self, doas, channels, levels, segment, orientation):
train_data = SplitData(doas_train, [ch_train], ['light', 'deep'], [segment], ['all'])
test_data = SplitData(doas_test, [ch_test], ['light', 'deep'], [segment], ['all'])
X_train, y_train = obtain_features_labels(train_data, encoding)
x_test, y_test = obtain_features_labels(test_data, encoding)
# model = SVC(gamma="auto")
# model = DecisionTreeClassifier(random_state=99, criterion='gini', max_depth=2)
model = RandomForestClassifier(n_estimators=5000, max_depth=5, min_samples_split=5, min_samples_leaf=10)
model.fit(X_train, y_train)
predictions = model.predict(x_test)
report = classification_report(y_test, predictions, output_dict=True)
acc = report['accuracy']
f1sc = report['weighted avg']['f1-score']
accuracies[ind_train][ind_test].append(acc)
f1scores[ind_train][ind_test].append(f1sc)
np.savetxt(write_file, np.array(ind_test), fmt="%s", newline=' ')
write_file.write('\n')
for ind_segment, segment in enumerate(segments):
for channel in range(len(all_channels)):
print("start running for channel " + str(all_channels[channel]) +
' ' + segment + '\n')
# SplitData(self, doas, channels, levels, segment, orientation):
train_data = ExtractData(doas_train, [all_channels[channel]],
['light', 'deep'], [segment], ['all'])
test_data = ExtractData(doas_test, [all_channels[channel]],
['light', 'deep'], [segment], ['all'])
X_train, y_train = obtain_features_labels(train_data,
encoding,
selected_symbols=15)
x_test, y_test = obtain_features_labels(test_data,
encoding,
selected_symbols=15)
model = RandomForestClassifier(n_estimators=5000,
max_depth=5,
min_samples_split=5,
min_samples_leaf=10)
model.fit(X_train, y_train)
predictions = model.predict(x_test)
report = classification_report(y_test,
predictions,
output_dict=True)